RU2816297C2 - Method and system for controlling web page field - Google Patents

Abstract

FIELD: physics.

SUBSTANCE: invention relates to computer engineering. Technical result is achieved due to the steps of obtaining, from a web server, data representing a web page for display in a browser application; determining, based on said data, web page elements associated with at least one third party web resource of the usage step; obtaining usage step data containing at least data on past user actions of the user with at least one third party web resource of the usage step; inputting the usage step data into the MLA algorithm to determine the probability value of allowing the user to use the third party cookie from at least one third party web resource of the usage step together with said web resource when browsing the web page; and determining the unlikely probability of allowing the user to use the third party cookie while browsing the web page if the probability value is less than the threshold value.

EFFECT: high protection of information in a computing device of a user against interference of third parties.

36 cl, 9 dwg

Description

Field of technology to which the invention relates

[01] The present technology generally relates to web browsing, and in particular to a method and system for managing tracking of web browsing activities in a browser.

State of the art

[02] When a user browses web pages on a network such as the Internet using a browser application such as Yandex.Browser™ browser application, Google Chrome™ browser application, Opera™ browser application, etc., some web resources can track his web browsing activities, such as clicking links or clicking buttons, starting or logging out of his user accounts, typing text data, etc. To this end, web resources may cause the browser application to receive and install cookies on the user device on which the application is running, also sometimes called web cookies, Internet cookies or browser cookies, i.e. . blocks of data specially generated to provide web servers of such web resources with the ability to track user browsing operations.

[03] For example, a cookie may be sent by the web resource whose web page the user is currently viewing (also called a first-party web page), i.e. The domain attribute of such a cookie corresponds to the domain name specified in the address bar of the browser application. This cookie is usually called a first-party cookie.

[04] However, there are another type of cookies, called third-party cookies, whose domain attributes are different from the domain name of the web page currently being viewed, indicated in the address bar of the browser application. Typically, third party cookies are delivered to the user's device by third party web properties associated with the first party web page the user is currently viewing, for example through certain layout elements. In particular, the third party web resource may be a social network (for example, VK.COM™) and may be linked to the first party web page through a “Like” or “Share” button. or through a comment field associated with the content of the first party's web page. In another example, a third party web property may be an e-commerce platform that has a commercial partnership with the first party's web page and may therefore be linked to it through an advertising banner or link that redirects the user to that platform's web page for e-commerce.

[05] Third party cookies sent in this manner may be received by a browser application and used to transmit data about the user's browsing activity to the relevant third party web server of the third party web property, which may pose risks to the user's personal information on the Internet . The third party server may also use browsing activity data for commercial purposes, such as providing user-oriented (targeted) online advertisements, etc., which may negatively impact user satisfaction.

[06] To solve the technical problem described above, some well-known approaches have been proposed to block third party cookies.

[07] Patent US10778792B1, “Providing user control of tracking user behavior” (International Business Machines Corp, issued September 15, 2020), describes a method, system, and computer program product for providing user control over tracking user behavior. It detects a request from a tracker (such as a website) to install content (such as tracking cookies) on the user's computing device for the purpose of tracking user behavior. In another embodiment, installation of a content tracker on a user's computing device is detected for the purpose of tracking user behavior. If the user has previously specified standard terms of service that the tracking tool must adhere to in order to track the user's behavior, it is determined whether the tracking tool agrees to accept such standard terms of service. If the tracking tool does not agree to accept such standard terms of service, then the tracking tool is blocked from tracking the user's online behavior, including blocking the tracker from installing content on the user's computing device for the purpose of tracking the user's behavior.

[08] Patent CN108124014B “Method for intelligently preventing third-party cookie tracking of browser” (Casic Wisdom Industrial Development Co LTD, issued September 22, 2020) describes a method for a browser to intelligently prevent third-party cookie tracking. This method allows the browser to restrict third-party cookies and website data, intelligently delete relevant cookie information, prevent cross-site tracking from cookies, and protect the user's sensitive data during online activities. Thus, by implementing the method according to an embodiment of this invention, the user does not need to manually clear the remaining cookie data every time after browsing a web page, and the browser can perform intelligent data clearing for the user. The browser may only store cookie data of a website that the user actually interacts with over an extended period of time, and identified cookie data may be deleted. As a result, the user’s personal data is reliably protected from theft.

Disclosure of the Invention

[09] In connection with the above, systems and methods are required to avoid, reduce or overcome the limitations of known technologies.

[010] The developers of this technology have determined that a user can be advised to accept or reject third party cookies based on the user's past user activity on the third party web resource. In particular, developers have developed non-limiting embodiments of methods and systems for training a machine learning algorithm (MLA) to determine, based on at least past user interactions with a third party web resource, what a user is likely to , will accept or reject third party cookies from a third party web property. In other words, the present methods and systems are designed to train an MLA algorithm to determine, based on past user actions, whether a third party web resource is perceived by the user as trustworthy or untrustworthy for receiving third party cookies from it.

[011] Additionally, in at least some embodiments of the present technology, these methods may include generating an appropriate notification advising the user to accept or reject a third party cookie from the third party web property. For example, in some non-limiting embodiments of the present technology, a user may be prompted to make a decision regarding whether to accept or reject a third party cookie from a third party web property only in response to a determination that the user is likely to reject the third party cookie. sides.

[012] Thus, non-limiting embodiments of the present technology are intended to implement a user-centric approach to accepting or rejecting third party cookies, thereby improving the user's browsing experience and improving the security of their personal information.

[013] In particular, according to the first aspect of the present technology, a computer method for managing tracking of a target user's web browsing operations in a browser application is implemented. The method is implemented by an electronic device running a browser application. The electronic device is connected to a server that executes the MLA algorithm. The MLA algorithm is trained by the server to determine the probability values for a usage stage user to allow third party cookies from third party web properties of the usage stage in conjunction with third party web properties when viewing usage stage web pages containing elements associated with third party web properties sides of the use stage. The MLA algorithm is trained on a training dataset containing many training objects. The training object of the plurality of training objects contains: (a) third party training web resource-dependent data of the third party training web resource associated with the training web page; (b) training user-specific data of the training user viewing the training web page, including data about the training user's past user activity on a third party training web resource; and (c) a label indicating whether or not the training user is allowed to use a third party training cookie from the third party training web resource in conjunction with the third party training web resource when viewing a training web page containing elements related to the training web - a third party resource, wherein the third party cookie was previously received by the training user's educational electronic device. The method includes: receiving by an electronic device from a given web server data representing a web page of a usage stage for display in a browser application; determining by the electronic device, based on data representing the use stage web page, elements thereof associated with at least one third party web resource of the use stage, wherein such element is capable of sending data about the actions thereof to a third party web server containing at least a third party web resource other than this web server; receipt by an electronic device of use-phase data containing (a) user-specific data of a target user, including data about the target user's past user activities with at least one third-party use-phase web property, and/or (b) web-specific data third party data of at least one third party web resource of the usage stage; input by an electronic device of use-phase data into the MLA algorithm to determine a probability value of the target user allowing the use of a third-party use-phase cookie from at least one third-party use-phase web property in conjunction with at least one third-party use-phase web property while browsing a Usage Stage web page, wherein a third party Usage Stage cookie has previously been received by the target user's electronic device and enables the third party server to track the target user's web browsing activities in the browser application and cookie sharing permission probability values -a third-party use-phase file from at least one third-party web resource of the use-phase for the target user and for the other target user are different; determining by the electronic device that it is unlikely that the target user will consent to the use of a third party cookie of the use step in conjunction with at least one third party web property of the use step while viewing a web page of the use step if the probability value is less than a probability threshold; and generating a first notification by the electronic device advising the target user not to use the use-phase third party cookie in conjunction with the at least one use-phase third party web property.

[014] In some embodiments of the method, if the probability value is less than a probability threshold, the method further includes causing the electronic device to prevent the browser application from sharing the third party cookie with the at least one third party web resource in an exploit step.

[015] In some embodiments of the method, if the probability value is greater than or equal to a probability threshold value, the method further includes: determining by the electronic device the target user's likely permission to use the third party cookie in conjunction with at least one third party web property sides of the use stage; and generating by the electronic device a second notification advising that it is safe to use the third party cookie in conjunction with at least one third party web resource of the use phase.

[016] In some embodiments of the method, it further includes: receiving by the electronic device an indication of a user action of the target user performed in response to the first notification or the second notification; and storing by the electronic device an indication of the user action for the purpose of updating the training data set for training the MLA algorithm if the user action does not correspond to the first notification or the second notification.

[017] In some embodiments of the method, if the probability value is greater than or equal to a probability threshold, the method further includes causing the electronic device to share a third party cookie from the at least one third party web resource with the browser application in a use step.

[018] In some embodiments of the method, the third party learning web resource data comprises a domain zone of the third party learning web resource and/or a domain name of the third party learning web resource and/or a category of the third party learning web resource.

[019] In some embodiments of the method, the following categories of third party web educational resource are used: (a) social network and/or (b) news portal and/or (c) video hosting platform.

[020] In some embodiments of the method, the user-specific data of the target user further includes data about the current location of the target user.

[021] In some embodiments of the method, the server is associated with another user's non-electronic device running another browser application. Another electronic device is capable of: receiving data from a web server representing a use stage web page for display in another browser application; determine, based on data representing the usage stage web page, elements thereof associated with at least one third party web resource of the usage stage; obtain other usage stage data containing (a) other user-dependent data of another user, including data about another target user's past user activities with at least one usage stage third party web property, and/or (b) web-specific third party resource data of at least one third party web resource of the use stage; inputting other usage phase data into the MLA algorithm to determine a different probability value of allowing another target user to use a usage phase third party cookie in conjunction with at least one usage phase third party web property while viewing a usage phase web page, wherein A third-party usage-stage cookie has previously been received by another target user's other electronic device and provides the third-party server with the ability to track the other user's web browsing activity in a different browser application and a different probability of allowing the other target user to use the third-party cookie. the use stage party in conjunction with at least one third party web resource of the use stage is different from the corresponding probability value determined for the target user; determine the likelihood of another target user authorizing the use of a third party cookie of the use step in conjunction with at least one third party web resource of the use step while viewing a web page of the use step if the other probability value is less than a probability threshold; and generate a first notice advising the other target user to opt out of using the use-phase third-party cookie in conjunction with at least one use-phase third-party web property.

[022] In some embodiments of the method, the MLA algorithm is a CatBoost decision tree MLA algorithm.

[023] According to a second aspect of the present technology, a computer method is implemented for managing tracking of a user's web browsing operations in a browser application. The method is implemented by the processor of an electronic device running a browser application. The method includes: receiving by a processor from a given web server data representing a web page for display in a browser application; determining by the processor, based on data representing the web page, elements thereof associated with at least one third party web resource, wherein such element is capable of sending data about actions thereon to a third party web server containing at least one web resource third party and other than this web server; the processor's receipt of (a) user-specific user data, including data about the user's past user activities with at least one third party web resource; and (b) third party web property-dependent data of at least one third party web property; input by the processor of user-specific data and third-party web resource-dependent data into the MLA algorithm to determine the probability value of the user allowing the use of a third-party cookie in conjunction with at least one third-party web resource while browsing a web page, when In this case, the third-party usage-stage cookie is previously received by the user's electronic device, and the third-party cookie enables the third-party server to track the user's web browsing activity in the browser application, and the MLA algorithm is trained to determine the probability values of the user's permission to use cookies third parties together with third party web resources when viewing related web pages based on at least user-specific data; wherein the probability values for allowing the use of a third party cookie in conjunction with at least one third party web resource are different for a user and for another user; determining that the processor is unlikely to allow the user to use a third party cookie in conjunction with at least one third party web resource of the usage step while browsing a web page if the probability value is less than a probability threshold; and generating a first notification by the processor advising the user not to use the third party cookie in conjunction with the at least one third party web property.

[024] In some embodiments of the method, if the probability value is less than a probability threshold, the method further includes causing the processor to prevent the browser application from sharing the third party cookie with the at least one third party web resource.

[025] In some embodiments of the method, if the probability value is greater than or equal to a probability threshold value, the method further includes:

determining by the processor the user's likely permission to use the third party cookie in conjunction with at least one third party web property; and generating a second notification by the processor advising that the third party cookie is safe to use in conjunction with the at least one third party web resource.

[026] In some embodiments of the method, it further includes: the processor receiving an indication of a user user action performed in response to the first notification or the second notification; and storing by the processor an indication of a user action associated with the at least one third party web resource for the purpose of updating a training dataset used to train the MLA algorithm if the user action does not correspond to the first notification or the second notification.

[027] In some embodiments of the method, if the probability value is greater than or equal to a probability threshold, the method further includes causing the processor to use the third party cookie in conjunction with the at least one third party web resource by the browser application.

[028] In some embodiments of the method, the user-specific data comprises (a) data about the user's current location and/or (b) data about the user's past user activities with at least one third party web resource.

[029] In some embodiments of the method, data about a user's past user activity on at least one third party web property comprises (a) the number of times the user visited the at least one third party web property during a predetermined period and/or (b ) the total number of visits by a user to at least one third party web resource.

[030] In some embodiments of the method, the third party web resource dependent data comprises (a) a domain zone of at least one third party web resource and/or (b) a domain name of at least one third party web resource and/ or (c) the category of at least one third party web property.

[031] In some embodiments of the method, the following categories of third party web educational resource are used: (a) social network and/or (b) news portal and/or (c) video hosting platform.

[032] In some embodiments of the method, the MLA algorithm is trained by a server associated with the electronic device based on a training data set. The training dataset contains many training objects. The learning object contains: (a) third party learning web resource-dependent third party learning web resource data associated with the learning web page; (b) training user-dependent data of the training user viewing the training web page, including consent or opt-out of receiving data from the training user's past user activities with a third party training web resource; and (c) a label indicating whether or not the training user is allowed to use a third party training cookie from the third party training web resource in conjunction with the third party training web resource when viewing a training web page containing elements related to the training web - a third party resource, wherein the third party cookie was previously received by the training user's educational electronic device.

[033] In some embodiments of the method, the third party learning web resource data comprises a third party learning web resource domain zone and/or a third party learning web resource domain name and/or a third party learning web resource category.

[034] In some embodiments of the method, the user and another user view the use step web page from different electronic devices.

[035] In some embodiments of the method, the MLA algorithm is a CatBoost decision tree MLA algorithm.

[036] According to the third aspect of the present technology, a computer method is implemented for training an MLA algorithm to determine the probability values of a target user's permission to use third party cookies from third party web resources of the usage stage together with third party web resources when browsing web pages of the usage stage . Usage Stage web pages contain elements associated with third party Usage Stage web properties. The method is implemented by a server connected to a plurality of educational electronic devices. The method includes: receiving by a server from a plurality of educational electronic devices a plurality of training objects, each of which contains: (a) third party training web resource-dependent data of a third party training web resource associated with the training web page, wherein the training the third party web resource and the training web page are hosted on different web servers; (b) training user-specific data of the training user viewing the training web page, including data about the training user's past user activity on a third party training web resource; and (c) a label indicating whether or not the training user is allowed to use a third party training cookie from the third party training web resource in conjunction with the third party training web resource when viewing a training web page containing elements related to the training web - a third party resource, wherein the third party cookie was previously received by the training user's electronic learning device; inputting by the server a plurality of training objects into the MLA algorithm and thereby determining an MLA algorithm formula for probability values of target users allowing the use of a third party cookie in conjunction with a third party web resource of the use phase; and sending the server an MLA algorithm formula to at least one target electronic device of the target user to determine a probability value of the target user's permission to use the third party cookie in conjunction with the third party web resource of the use step, wherein the probability value of the target user permission to use the cookie is third party file in conjunction with a third party web resource, the usage phase differs for different target users.

[037] In some embodiments of the method, the third party training web resource data comprises a domain zone of the third party training web resource and/or a domain name of the third party training web resource and/or a category of the third party training web resource.

[038] In some embodiments of the method, the following categories of third party web educational resource are used: (a) social network and/or (b) news portal and/or (c) video hosting platform.

[039] In some embodiments of the method, the MLA algorithm is a CatBoost decision tree MLA algorithm.

[040] According to a fourth aspect of the present technology, an electronic device is implemented for managing tracking of a target user's web browsing operations in a browser application of the electronic device. The electronic device is connected to a server that executes the MLA algorithm. The MLA algorithm is trained by the server to determine the probability values for a usage stage user to allow third party cookies from third party web properties of the usage stage in conjunction with third party web properties when viewing usage stage web pages containing elements associated with third party web properties sides of the use stage. The MLA algorithm is trained on a training dataset containing many training objects. The training object of the plurality of training objects contains: (a) third party training web resource-dependent data of the third party training web resource associated with the training web page; (b) training user-specific data of the training user viewing the training web page, including data about the training user's past user activity on a third party training web resource; and (c) a label indicating whether or not the training user is allowed to use a third party training cookie from the third party training web resource in conjunction with the third party training web resource when viewing a training web page containing elements related to the training web - a third party resource, wherein the third party cookie was previously received by the training user's educational electronic device. The electronic device contains a processor and a machine-readable physical storage medium that stores instructions. The processor, when executing commands, is capable of: receiving from a given web server data representing the web page of the use stage for display in a browser application; determine, based on data representing the use-phase web page, elements thereof associated with at least one third-party use-phase web resource, wherein such element is capable of sending activity data thereon to a third party web server containing at least a third party web resource other than this web server; obtain usage stage data containing (a) user-specific data of the target user, including data about the target user's past user activities with at least one third party web property of the usage stage, and/or (b) specific to the third party web property data from at least one third party web resource of the usage stage; input the usage phase data into the MLA algorithm to determine a probability value of the target user allowing a usage phase third party cookie from at least one usage phase third party web property in conjunction with at least one usage phase third party web resource during usage-stage web page browsing, wherein the third-party usage-stage cookie has previously been received by the target user's electronic device and enables the third party server to track the target user's web browsing activities in the browser application and the cookie sharing permission probability values third party use phase from at least one third party web resource the use phase for the target user and for the other target user are different; determine the likelihood of the target user agreeing to use the use step third party cookie in conjunction with at least one use step third party web property while viewing the use step web page if the probability value is less than a likelihood threshold; and generate a first notice advising the target user to opt out of using the third party usage stage cookie in conjunction with at least one third party usage stage web property.

[041] In some embodiments of the electronic device, if the probability value is less than a probability threshold, the processor is further capable of causing the browser application to prevent the third party cookie from being shared with the at least one third party web resource of the usage step.

[042] In some embodiments of the electronic device, if the probability value is greater than or equal to a probability threshold, then the processor is further capable of: determining the target user's likely permission to use the third party cookie in conjunction with at least one third party web resource of the step use; and generate a second notice advising that it is safe to use the third party cookie in conjunction with at least one third party web resource of the use phase.

[043] According to a fifth aspect of the present technology, a computer method for controlling tracking of a target user's web browsing operations in a browser application is implemented. The method is implemented by an electronic device running a browser application. The electronic device is connected to a server that executes the MLA algorithm. The MLA algorithm is trained by the server, based on a training dataset containing a plurality of training objects, to determine probability values for a use-phase user to allow third-party cookies from use-phase third-party web resources when viewing associated use-phase web pages. The training object of the plurality of training objects contains: (a) third party training web resource-dependent data of the third party training web resource associated with the training web page; (b) training user-specific data of the training user viewing the training web page, including data about the training user's past user activity on a third party training web resource; and (c) a label indicating whether or not the training user is allowed to receive a third party training cookie from the third party training web resource when viewing the training web page. The method includes: receiving by an electronic device from a given web server data representing a web page of a usage stage for display in a browser application; determining by the electronic device, based on data representing the use stage web page, elements thereof associated with at least one third party web resource of the use stage, wherein such element is capable of sending data about the actions thereof to a third party web server containing at least a third party web resource other than this web server; receipt by an electronic device of use-phase data containing (a) user-specific data of a target user, including data about the target user's past user activities with at least one third-party use-phase web property, and/or (b) web-specific data third party data of at least one third party web resource of the usage stage; input by an electronic device of use-phase data into the MLA algorithm to determine a probability value of the target user allowing the receipt of a third-party use-phase cookie from at least one third-party use-phase web resource while viewing a use-phase web page, wherein the cookie the third-party use-phase file enables the third-party server to track a target user's web browsing activity in a browser application, and the probability values of allowing a third-party use-phase cookie from at least one third-party use-phase web resource for the target user, and for another target user are different; determining by the electronic device that it is unlikely that the target user will authorize the receipt of a third party cookie of the use step from at least one third party web resource of the use step while viewing a web page of the use step if the probability value is less than a probability threshold; and generating a first notification by the electronic device advising the target user to opt-out of receiving a third-party usage-stage cookie from the at least one third-party usage-stage web property.

[044] In some embodiments of the method, if the probability value is less than a probability threshold, the method further includes causing the electronic device to refuse the browser application to receive a third party cookie from the at least one third party web resource in the exploit step.

[045] In some embodiments of the method, if the probability value is greater than or equal to a probability threshold, the method further includes: determining by the electronic device the target user's likely permission to receive a third party cookie from the at least one third party web resource stage of use; and generating, by the electronic device, a second notification indicating security permission to receive the third party cookie from the at least one third party web resource of the use phase.

[046] In some embodiments of the method, it further includes: receiving by the electronic device an indication of a user action of the target user performed in response to the first notification or the second notification; and storing by the electronic device an indication of the user action for the purpose of updating the training data set for training the MLA algorithm if the user action does not correspond to the first notification or the second notification.

[047] In some embodiments of the method, if the probability value is greater than or equal to a probability threshold value, the method further includes causing the electronic device to allow the browser application to receive a third party cookie from the at least one third party web resource in an exploit step .

[048] In some embodiments of the method, the third party learning web resource data comprises a third party learning web resource domain zone and/or a third party learning web resource domain name and/or a third party learning web resource category.

[049] In some embodiments of the method, the following categories of third party web educational resource are used: (a) social network and/or (b) news portal and/or (c) video hosting platform.

[050] In some embodiments of the method, the user-specific target user data further includes data about the current location of the target user.

[051] In some embodiments of the method, the server is associated with another electronic device of another target user other than the electronic device. Another electronic device is running another browser application. Another electronic device is capable of: receiving data from a web server representing a use stage web page for display in another browser application; determine, based on data representing the usage stage web page, elements thereof associated with at least one third party web resource of the usage stage; obtain other usage stage data containing (a) other user-dependent data of another user, including data about another target user's past user activities with at least one usage stage third party web property, and/or (b) web-specific third party resource data of at least one third party web resource of the use stage; input other usage stage data into the MLA algorithm to determine a different probability value of allowing another target user to receive a usage stage third party cookie from at least one usage stage third party web resource while browsing a usage stage web page, wherein the cookie -the third party use stage file provides the third party server with the ability to track web browsing activity performed by another user in another browser application, and another value of the probability of allowing the receipt of the third party use stage cookie from at least one third party stage web resource usage for another target user is different from the probability value determined for the target user; determine the likelihood of another target user allowing the use stage third party cookie from at least one use stage third party web resource while browsing the use stage web page if the other probability value is less than a probability threshold; and generating a first notice advising the other target user to opt-out of receiving a third-party usage-stage cookie from the at least one third-party usage-stage web property.

[052] In some embodiments of the method, the MLA algorithm is a CatBoost decision tree MLA algorithm.

[053] According to a sixth aspect of the present technology, a computer method for managing tracking of a user's web browsing operations in a browser application is implemented. The method is implemented by the processor of an electronic device running a browser application. The method includes: receiving by a processor from a given web server data representing a web page for display in a browser application; determining by the processor, based on data representing the web page, elements thereof associated with at least one third party web resource, wherein such element is capable of sending data about actions thereon to a third party web server containing at least one web resource third party and other than this web server; the processor's receipt of (a) user-specific user data, including data about the user's past user activities with at least one third party web resource; and (b) third party web property-dependent data of at least one third party web property; input by the processor of user-specific data and third-party web resource-dependent data into the MLA algorithm to determine a probability value of the user's permission to receive a third-party cookie from at least one third-party web resource while browsing a web page, wherein a third party cookie enables a third party server to track a user's web browsing activity in a browser application, and the MLA algorithm is trained to determine probability values for the user to allow third party cookies from third party web resources based on at least data user; wherein the values of the probability of permission to receive a third party cookie from at least one third party web resource for the user and for the other user are different; determining that the processor is unlikely to allow the user to receive a third party cookie from the at least one third party web resource of the usage step while browsing a web page if the probability value is less than a probability threshold; and generating a first notification by the processor advising the user not to receive a third party cookie from the at least one third party web resource.

[054] In some embodiments of the method, if the probability value is less than a probability threshold, the method further includes causing the processor to refuse to receive the third party cookie from the at least one third party web resource.

[055] In some embodiments of the method, if the probability value is greater than or equal to a probability threshold, the method further includes: determining by the processor the user's likely permission to receive a third party cookie from the at least one third party web resource; and generating a second notification by the processor indicating security permission to receive the third party cookie from the at least one third party web resource.

[056] In some embodiments of the method, it further includes: the processor receiving an indication of a user user action performed in response to the first notification or the second notification; and storing by the processor an indication of a user action associated with the at least one third party web resource for the purpose of updating a training dataset used to train the MLA algorithm if the user action does not correspond to the first notification or the second notification.

[057] In some embodiments of the method, if the probability value is greater than or equal to a probability threshold, the method further includes causing the processor to allow the browser application to receive a third party cookie from the at least one third party web resource

[058] In some embodiments of the method, the user-specific data comprises (a) data about the user's current location and/or (b) data about the user's past user activities with at least one third party web resource.

[059] In some embodiments of the method, data about a user's past user activity on at least one third party web property comprises (a) the number of times the user visited the at least one third party web property during a predetermined period and/or (b ) the total number of visits by a user to at least one third party web resource.

[060] In some embodiments of the method, the third party web resource dependent data comprises (a) a domain zone of at least one third party web resource and/or (b) a domain name of at least one third party web resource and/ or (c) the category of at least one third party web property.

[061] In some embodiments of the method, the following categories of third party educational web resource are used: (a) social network and/or (b) news portal and/or (c) video hosting platform.

[062] In some embodiments of the method, the MLA algorithm is trained by a server associated with the electronic device based on a training data set containing a plurality of training objects, each of which contains: (a) third party training web resource-dependent training web resource data a third party associated with the training web page; (b) training user-specific data of the training user viewing the training web page, including data about the training user's past user activity on a third party training web resource; and (c) a label indicating whether or not the training user is allowed to receive a third party training cookie from the third party training web resource when viewing the training web page.

[063] In some embodiments of the method, the third party training web resource data comprises a domain zone of the third party training web resource and/or a domain name of the third party training web resource and/or a category of the third party training web resource.

[064] In some embodiments of the method, the user and another user view the use step web page from different electronic devices.

[065] In some embodiments of the method, the MLA algorithm is a CatBoost decision tree-based MLA algorithm.

[066] According to the seventh aspect of the present technology, a computer method is implemented for training an MLA algorithm to determine a probability value of a target user's permission to receive a third party cookie when viewing a usage stage web page. The method is implemented by a server connected to a plurality of educational electronic devices. The method includes: receiving by a server from a plurality of educational electronic devices a plurality of training objects, each of which contains: (a) third party training web resource-dependent data of a third party training web resource associated with the training web page, wherein the training the third party web resource and the training web page are hosted on different web servers; (b) training user-specific data of the training user viewing the training web page, including data about the training user's past user activity on a third party training web resource; and (c) a label indicating whether the training user is permitted or not allowed to receive a third party training cookie from the third party training web resource when viewing the training web page; inputting by the server a plurality of training objects into the MLA algorithm and thereby determining an MLA algorithm formula for probability values of allowing target users to receive a third party cookie from a third party web resource of the use phase; and sending the server an MLA algorithm formula to at least one target electronic device of the target user to determine a probability value of allowing the target user to receive a third party cookie from the third party web resource of the use step, wherein a probability value of allowing the target user to receive the cookie third party versus third party web resource, the usage stages differ for different target users.

[067] In some embodiments of the method, the third party training web resource data comprises a domain zone of the third party training web resource and/or a domain name of the third party training web resource and/or a category of the third party training web resource.

[068] In some embodiments of the method, the following categories of third party educational web resource are used: (a) social network and/or (b) news portal and/or (c) video hosting platform.

[069] In some embodiments of the method, the MLA algorithm is a CatBoost decision tree MLA algorithm.

[070] According to the eighth aspect of the present technology, an electronic device is implemented for managing tracking of a target user's web browsing operations in a browser application of the electronic device. The electronic device is connected to a server that executes the MLA algorithm. The MLA algorithm is trained by the server, based on a training dataset containing a plurality of training objects, to determine probability values for a use-phase user to allow third-party cookies from use-phase third-party web resources when viewing associated use-phase web pages. The training object of the plurality of training objects contains: (a) third party training web resource-dependent data of the third party training web resource associated with the training web page; (b) training user-specific data of the training user viewing the training web page, including data about the training user's past user activity on a third party training web resource; and (c) a label indicating whether or not the training user is allowed to receive a third party training cookie from the third party training web resource when viewing the training web page. The electronic device contains a processor and a machine-readable physical storage medium that stores instructions. The processor, when executing commands, is capable of: receiving from a given web server data representing the web page of the use stage for display in a browser application; determine, based on data representing the use-phase web page, elements thereof associated with at least one third-party use-phase web resource, wherein such element is capable of sending activity data thereon to a third party web server containing at least a third party web resource other than this web server; obtain usage stage data containing (a) user-specific data of the target user, including data about the target user's past user activities with at least one third party web property of the usage stage, and/or (b) specific to the third party web property data from at least one third party web resource of the usage stage; input the usage stage data into the MLA algorithm to determine a probability value of the target user allowing a third party usage stage cookie from at least one third party usage stage web resource while viewing a usage stage web page, wherein the third party cookie The use-stage side provides the third party server with the ability to track a target user's web browsing activity in a browser application, and the likelihood values of allowing a third-party use-stage cookie from at least one use-stage third party web resource for the target user and for another target user are different; determine whether the target user is unlikely to allow the use stage third party cookie from at least one use stage third party web resource while browsing the use stage web page if the probability value is less than a probability threshold; and generating a first notice advising the target user to opt-out of receiving a third-party usage-stage cookie from the at least one third-party usage-stage web property.

[071] In some embodiments of the electronic device, if the probability value is less than a probability threshold, the processor is further capable of causing the browser application to refuse to receive a third party cookie from the at least one third party web resource of the usage stage.

[072] In some embodiments of an electronic device, if the probability value is greater than or equal to a probability threshold, the processor is further capable of:

[073] determine the target user's likely permission to receive a third party cookie from at least one third party web resource of the use phase; and generate a second notification indicating the security of permission to receive the third party cookie from the at least one third party web resource of the use phase.

[074] In some embodiments of the electronic device, the processor is further capable of:

[075] receive an indication of a target user's custom action performed in response to the first notification or the second notification; and store on a computer-readable physical storage medium an indication of the user action for the purpose of updating a training data set for training the MLA algorithm if the user action does not correspond to the first notification or the second notification.

[076] As used herein, the term “server” means a computer program executed by associated hardware and capable of receiving requests (eg, from electronic devices) via a network and executing or initiating execution of those requests. The hardware may be one physical computer or one computer system, which is not essential for this technology. In the present context, the expression "server" does not mean that every task (for example, a received command or request) or any particular task is accepted, executed or launched by the same server (i.e. the same software and/or hardware ). This expression means that any number of software or hardware can receive, send, execute, or initiate execution of any task or request, or the results of any tasks or requests. All of these software and hardware may be a single server or multiple servers, both of which are implied by the expression “at least one server.”

[077] As used herein, the term "electronic device" means any computer hardware capable of executing programs suitable for solving a given task. As used herein, the term "electronic device" implies that the device can function as a server for other electronic devices, however, this is not necessary for the present technology. Thus, some non-limiting examples of electronic devices include self-driving units, personal computers (desktops, laptops, netbooks, etc.), smartphones and tablets, and network equipment such as routers, switches, and gateways. It should be understood that in the present context, the fact that a device functions as an electronic device does not mean that it cannot function as a server for other electronic devices.

[078] As used herein, the expression “information” includes information of any kind or kind capable of being stored in a database. Thus, information includes visual works (for example, maps), audio-visual works (for example, images, films, sound recordings, presentations, etc.), data (for example, location data, meteorological data, traffic data, numerical data, etc.), text (such as, but not limited to, opinions, comments, questions, messages, etc.), documents, spreadsheets, etc.

[079] As used herein, the term “database” means any structured collection of data, regardless of its specific structure, database management software, or computer hardware for storing, using, or otherwise making use of that data. The database may reside in the same hardware as the process that stores or uses the information stored in the database, or the database may reside in separate hardware, such as a dedicated server or multiple servers.

[080] In the context of the present description, the numerals “first”, “second”, “third”, etc. are used only to indicate differences between the nouns they refer to, but not to describe any specific relationships between those nouns. Moreover, as occurs herein in other contexts, reference to a “first” element and a “second” element does not preclude the fact that the two elements may in fact be the same element.

[081] Each embodiment of the present technology relates to at least one of the above-mentioned objectives and/or one of the above-mentioned aspects, but not necessarily all of them. It should be understood that some aspects of the present technology associated with attempting to achieve the above purpose may not be consistent with that purpose and/or may be consistent with other purposes not expressly mentioned herein.

[082] Additional and/or alternative features, aspects and advantages of embodiments of the present technology are set forth in the following description, the accompanying drawings, and the claims.

Brief description of drawings

[083] These and other features, aspects and advantages of the present technology are set forth in the following description, in the appended claims and in the following drawings.

[084] In FIG. 1 is a diagram of an example computer system that may be configured to implement certain non-limiting embodiments of the present technology.

[085] In FIG. 2 is a diagram of a networked computing environment containing the computer system shown in FIG. 1, and suitable for use with certain non-limiting embodiments of the present technology.

[086] In FIG. 3 is a schematic representation of a graphical user interface (GUI) of a browser application running on an electronic device in a networked computing environment shown in FIG. 2, and providing a visual representation of a web page associated with a third party web resource, in accordance with certain non-limiting embodiments of the present technology.

[087] In FIG. 4 is a block diagram of a machine learning model architecture used to implement at least some non-limiting embodiments of the present technology.

[088] In FIG. Figure 5 shows a diagram of the formation of a set of training digital objects for training the architecture of a machine learning model according to FIG. 4 determining the probability value of user acceptance of the electronic device presented in the network computing environment in FIG. 2, a third party cookie from a third party web resource when viewing the web page shown in FIG. 3, according to some non-limiting embodiments of the present technology.

[089] In FIG. 6 is a schematic view of the web page according to FIG. 3 containing a first notification generated by an electronic device represented in the network computing environment in FIG. 2, for the corresponding user based on the probability value determined using the trained machine learning model architecture presented in FIG. 4, according to some non-limiting embodiments of the present technology.

[090] In FIG. 7 is a schematic view of the web page according to FIG. 3 containing a second notification generated by an electronic device represented in the network computing environment in FIG. 2, for the corresponding user based on the probability value determined using the trained machine learning model architecture presented in FIG. 4, according to some non-limiting embodiments of the present technology.

[091] In FIG. 8 is a flowchart of a first method for training a machine learning model architecture according to FIG. 4 determining the probability value of user acceptance of the electronic device presented in the network computing environment in FIG. 2, a third party cookie from a third party web resource when viewing the web page shown in FIG. 3, according to some non-limiting embodiments of the present technology.

[092] In FIG. 9 is a flowchart of a second method for managing tracking of web browsing operations by a user of an electronic device represented in the network computing environment in FIG. 2, using the machine learning model architecture of FIG. 4, trained in accordance with the first method presented in FIG. 8, according to some non-limiting embodiments of the present technology.

Carrying out the invention

[093] The examples and conventional language presented here are intended to provide a better understanding of the principles of this technology and not to limit its scope to such specifically provided examples and conditions. It is obvious that those skilled in the art are capable of developing various methods and devices that are not explicitly described or shown, but implement the principles of the present technology within its spirit and scope.

[094] In addition, to facilitate a better understanding, the following description may contain simplified embodiments of the present technology. Those skilled in the art will appreciate that other embodiments of the present technology may be significantly more complex.

[095] In some cases, useful examples of modifications to the present technology are provided. They promote understanding, but also do not define the scope or boundaries of the actual technology. The presented list of modifications is not exhaustive and a person skilled in the art may develop other modifications within the scope of this technology. In addition, if in some cases modifications are not described, this does not mean that they are impossible and/or that the description contains the only possible implementation of a particular element of this technology.

[096] Moreover, the descriptions of the principles, aspects and embodiments of the present technology, as well as specific examples thereof, are intended to cover their structural and functional equivalents, whether currently known or developed in the future. For example, those skilled in the art will appreciate that any block diagrams described herein correspond to the conceptual representations of illustrative circuit diagrams implementing the fundamentals of the present technology. It should also be clear that any flowcharts, process diagrams, state transition diagrams, pseudo codes, etc. correspond to various processes that may be represented on a computer-readable physical storage medium and may be performed by a computer or processor, whether such computer or processor is explicitly shown or not.

[097] The functions of the various elements shown in the drawings, including any functional block designated as a “processor,” may be implemented using dedicated hardware, as well as using hardware capable of executing associated software. If a processor is used, these functions may be performed by one dedicated processor, one shared processor, and/or a plurality of separate processors, some of which may be shared. In addition, explicit use of the term "processor" or "controller" should not be construed as referring solely to hardware capable of executing software and may include, but is not limited to, digital signal processor (DSP) hardware, network processor, application specific integrated circuit (ASIC), field programmable gate array (FPGA), read-only memory (ROM) for storing software, random access memory (RAM), and/or non-volatile storage device. Other hardware, general purpose and/or custom, may also be included.

[098] Software modules, or simply modules intended to be implemented in software, may be represented herein as any combination of flowchart elements or other elements indicating the execution of process steps and/or containing a textual description. Such modules may be executed by hardware, shown explicitly or implicitly.

[099] Given the foregoing principles, some non-limiting examples are provided below to illustrate various embodiments of aspects of the present technology.

Computer system

[0100] In FIG. 1 illustrates a computer system 100 suitable for use in some embodiments of the present technology. The computer system 100 includes various hardware elements, including one or more single- or multi-core processors, generally represented by a processor 110, a graphics processing unit (GPU) 111, a solid-state drive 120, a RAM 130, a display interface 140, and an input/output interface 150.

[0101] Communication between various elements of the computer system 100 may occur through one or more internal and/or external buses 160 (such as a PCI bus, a USB bus, an IEEE 1394 FireWire bus, a SCSI bus, a Serial-ATA bus, etc. ) to which various hardware elements are connected electronically.

[0102] The input/output interface 150 may be coupled to the touch screen 190 and/or to one or more internal and/or external buses 160. The touch screen 190 may also be referred to as a screen (e.g., a screen (not specifically identified) of the first electronic device 210, presented in Fig. 2). In the ones shown in FIGS. 1 embodiments of the invention, touch screen 190 includes touch media 194 (e.g., pressure-sensitive cells built into the display layer and allowing physical action between the user and the display to be detected) and a touch input/output controller 192 that provides communication with display interface 140 and /or with one or more internal and/or external buses 160. In some non-limiting embodiments of the present technology, I/O interface 150 may be coupled to a keyboard (not shown), mouse (not shown), or touch pad (not shown) that provide user interaction with the computer system 100 in addition to or instead of the touch screen 190.

[0103] It should be noted that in some non-limiting embodiments of the present technology, certain elements of the computer system 100 may be missing. For example, a keyboard and mouse (not separately shown) may be absent, particularly if the computer system 100 is implemented as a small electronic device such as (but not limited to) a smartphone.

[0104] In embodiments of the present technology, solid state drive 120 stores software instructions suitable for loading into RAM 130 and execution by processor 110 and/or graphics processor 111. The software instructions may, for example, be part of a library or application.

Network computing environment

[0105] In FIG. 2 is a diagram of a networked computing environment 200 suitable for use with certain non-limiting embodiments of the present technology. Network computing environment 200 includes a first electronic device 210 and a second electronic device 212 connected via a communication network 240 to a first party web server 250. In some non-limiting embodiments of the present technology, the first electronic device 210 and the second electronic device 212 may be associated with users such as the first user 211 and the second user 213, respectively.

[0106] In non-limiting embodiments of the present technology, the first electronic device 210 and the second electronic device 212 may be any computer hardware capable of executing programs suitable for solving a given task. Thus, some non-limiting examples of first electronic device 210 and second electronic device 212 may include personal computers (desktops, laptops, netbooks, etc.), smartphones, and tablets. The first electronic device 210 and the second electronic device 212 may comprise some or all of the elements of the computer system 100 shown in FIG. 1.

[0107] In some non-limiting embodiments of the present technology, the first party web server 250 may be implemented as a traditional computer server and may comprise some or all of the elements of the computer system 100 shown in FIG. 1. In one non-limiting example, first party web server 250 is implemented as a Dell™ PowerEdge™ server running the Microsoft™ Windows Server™ operating system, but it may also be implemented using any other suitable hardware, application software and/or firmware or a combination thereof. In the illustrated non-limiting embodiments of the present technology, the first party web server 250 is a single server. In other non-limiting embodiments of the present technology (not shown), the functions of the first party web server 250 may be distributed across multiple servers.

[0108] According to some non-limiting embodiments of the present technology, the first party web server 250 may contain a web resource 255. In general, the web resource 255 may contain and thus provide access to various types of information, such as texts, images , audio and video materials hosted on at least one web page (such as web page 302 shown in FIG. 3) that can be identified in the communications network 240, for example, using an appropriate universal resource locator (URL). Resource Locator) associated with at least one page. Examples of web resource 255 include a website such as a commercial website designed to advertise specific products, a news portal (such as RIA Novosti™ or CNN™), an online video service (for example, Kinopoisk.ru™) and so on.

[0109] Thus, in some non-limiting embodiments of the present technology, a user using his electronic device, such as the first user 211 using the first electronic device 210, may request access to a web page 302, for example, by submitting a corresponding URL -addresses in the address bar (not shown in FIG. 2) of the browser application 215 running on the first electronic device 210. In another example, the first user 211 may be redirected to a web page 302 from another web page containing a link to the web page 302 that has been selected by the first user 211. In response, the first party web server 250 may provide the first electronic device 210 with access to the web page 302 of the web resource 255 by sending a digital document representing it to the first electronic device 210 for provision to the first user 211.

[0110] As used herein, browser application 215 may be a software application executing on an electronic device, such as first electronic device 210, and capable of providing access to resources of network nodes, such as communications network 240, in response to receiving their network addresses. online. In a specific, non-limiting example, the browser application 215 may be implemented as the Yandex.Browser™ browser application provided by Yandex LLC (16 Lev Tolstogo St., Moscow, 119021, Russia).

[0111] In addition, the corresponding digital document representing the web page 302 may contain executable source code, for example, developed in HyperText Markup Language (HTML). In other non-limiting embodiments of the present technology, at least a portion of the executable source code may be developed in another programming language, such as the JavaScript programming language and/or the Cascading Style Sheets (CSS) programming language.

[0112] Upon receipt of this digital document, the first electronic device 210 may cause the browser application 215 to execute executable source code associated with the web page 302 and, accordingly, generate, for example, a renderer of the browser application 215, a visual representation thereof for presentation to the first user 211.

[0113] In FIG. 3 illustrates a GUI interface of a browser application 215 (see FIG. 2) providing a visual representation of a web page 302, in accordance with some non-limiting embodiments of the present technology.

[0114] As shown in FIG. 3, according to some non-limiting embodiments of the present technology, a web page 302 may contain a plurality of layout elements that define its visual presentation. For example, the plurality of layout elements may contain their own layout elements 304 of the web page 302, i.e. layout elements representing the main content of the web page 302 provided by the first party web server 250. However, in some non-limiting embodiments of the present technology, the web page 302 may contain third party elements 308. Specifically, in these embodiments, a third party element 309 of third party elements 308 may represent at least a portion of the content of a third party web resource 265 hosted on a third party web server 260 different from the first party web server 250. containing web resource 255. That is, the third party element 309 may be associated with a domain name of the third party web server 260 that is different from the first party domain name 306 of the first party web server 250 shown in FIG. 3 in the address bar (not separately designated) of the GUI interface of the browser application 215.

[0115] Meanwhile, as shown in FIG. 2, a third party web server 260 may also be connected to the communications network 240. It will be appreciated that, according to some non-limiting embodiments of the present technology, third party web server 260 may be implemented similar to first party web server 250 and, therefore, may also include some or all of the elements of the computer system 100 described above with reference to fig. 1.

[0116] To present content from a third party web server 260, in some non-limiting embodiments of the present technology, the third party element 309 of the web page 302 may be linked to the third party web resource 265. As used herein, the term “linked” means that the third party element 309 may (a) redirect the user 211 viewing the web page 302 to a corresponding web page of the third party web resource 265 in response to a predetermined action of the first user 211 with third party element 309, such as a “click” on it, and/or (b) send to the third party web server 260 specific data indicating actions on only the third party element 309 and/or on the entire web page 302.

[0117] There is no limitation on the data that third party element 309 can send to third party web server 260. Typically, this data depends on the type of third party web resource 265. In one example, where the third party web resource 265 is a social network (such as the VK.COM™ social network or the OK.RU™ social network), as shown in FIG. 3, the third party element 309 may be a "Like" button or a "Share" button associated with the social network, and therefore may send it data indicative of the response of the first user 211 viewing web page 302, regarding the content of the web page 302, such as a positive (“like”) or negative reaction and/or use or non-use of the “share” function regarding the content of the web page 302 after rating it. In another example (not shown), the third party web property 265 may be a video hosting platform (such as the YouTube™ video hosting platform), and the third party element 309 may be an embedded video player of the video hosting platform that may send it data indicative of the number of views video content played by the built-in video player. In another example, in which the third party web property 265 is a commercial web property where specific products, such as goods or services, are offered for sale, as further illustrated in FIG. 3, the third party element 309 may be an advertising banner capable of sending to the commercial web resource data about user interactions with it by the first user 211, such as the number of “clicks”, time spent on the advertising banner, number of hovers on the advertising banner, etc. P.

[0118] However, it should be understood that embodiments of third party element 309 are not limited to the examples presented above. There may be other elements associated with third party web properties that are capable of sending specific data indicating user actions to the relevant third party web servers. In certain non-limiting embodiments of the present technology, the third party element 309 may not even be visible to the first user 211, for example, may only be a pixel in the visual representation of the web page 302, but may still send data indicative of user actions with web page 302 to a third party web server 260 as described above.

[0119] Additionally, according to some non-limiting embodiments of the present technology, web resource 255 may use cookies. Generally, in the context of this document, a cookie corresponds to a piece of data sent to a user's electronic device, such as first electronic device 210, while the corresponding user is browsing a web resource, such as web page 302 of web resource 255. For example, a cookie the file may be sent to the first electronic device 210 by executing a corresponding portion of the executable source code of the web page 302 (such as a JavaScript portion capable of creating a cookie on the first electronic device 210) or by using HTTP protocol headers when requesting access to web content -page 302 or third party web resource 265, for example, in response to an action on third party element 309. The cookie may then send specific data to the web server that sent the cookie to the first electronic device 210.

[0120] In various non-limiting embodiments of the present technology, such data may include data about the web browsing activities of the first user 211, such as, but not limited to, (a) the web browsing history of the first user 211; (b) clicking on buttons or following links on various web pages; (c) search operations performed on search engines (such as the Yandex™ search engine); (d) starting or ending a session on various resources, such as web email services, social networks, etc. In additional non-limiting embodiments of the present technology, the data that may be sent by the cookie to the appropriate web server may further comprise data from the first electronic device 210, including, but not limited to, (a) the model of the first electronic device 210; (b) information about the operating system, such as versions, dates of last updates, etc.; (c) data from sensors of the first electronic device 210, such as GPS data; etc. It should be noted that the above types of data that can be sent to the relevant web server are for explanation purposes only. Other types of data are also possible without going beyond the boundaries of this technology.

[0121] Additionally, if the cookie's domain attribute matches the domain name of the web resource that the corresponding user is currently viewing, such as the first-party domain name 306 of the first-party web server 250, then the cookie is called a cookie. a first party cookie, otherwise the cookie is called a third party cookie. That is, the cookies sent to the first electronic device 210 by the first party web server 250 while the first user 211 is browsing the web page 302 are first party cookies (see FIG. 3). Conversely, cookies sent to the first electronic device 210 by a third party web server 260 while the first user 211 is browsing a web page 302 are third party cookies.

[0122] In some non-limiting embodiments of the present technology, a third party cookie may be sent to the first electronic device 210 by the third party web server 260, for example, in response to the first user 211's action on the third party element 309 of the web page 302 In this regard, it is expected that the receipt of the above data by the third party server 260 using a third party cookie may cause concern to the first user 211. The third party server 260 may use the data thus obtained for certain commercial purposes, such as serving advertisements , targeting the first user 211, for example, based on his web browsing activities, which may negatively impact the first user 211's browsing satisfaction.

[0123] In some non-limiting embodiments of the present technology, while parsing the executable source code of web page 302, browser application 215 may (a) determine that web page 302 contains a third party element 309 associated with third party web resource 265 party, and (b) generate a notification 310 notifying the first user 211 that the web page 302 may use third party cookies, such as a third party cookie from the third party web server 260. In addition, the browser application 215 may provide the first user 211 with the ability to configure its preferences regarding third party cookies received from all web resources associated with the web page 302, and in particular from the third party web server 260. In a non-limiting example shown in FIG. 3, upon notification 310, the browser application 215 may prompt the first user 211 to (a) allow all third party cookies sent using the web page 302, or (b) deny receiving all third party cookies sent using the web -302 pages, or (c) set preferences by selecting web servers from which to accept or reject third party cookies.

[0124] In other non-limiting embodiments of the present technology, the third party cookie may be received in advance by the first electronic device 210 from the third party server 260. For example, using the communications network 240, the first user 211 could view other web pages (not shown) containing items associated with the third party web resource 265, such as the third party item 309, or one of the web pages of the web resource 265 and may have already allowed the cookie to be received from the third party server 260 on the first electronic device 210. In this case, when the first user 211 views the web page 302, the third party server 260 may (1) determine that its third party cookie already received by the electronic device, and (2) send an additional request to share a third party cookie to continue tracking the web browsing activities of the first user 211 on the communication network 240. Upon receiving a request to share a third party cookie from the third party server 260, the browser application 215 may offer the first user 211 similar options as described above: (a) allowing the sharing of all third party cookies for the web page 302; (b) prohibit the sharing of all third party cookies for the 302 web page; and (c) selecting third party web properties with which third party cookies should or should not be used.

[0125] Some known approaches suggest that the default setting is for the browser application 215 to initiate blocking of any third party cookies (or sharing them). However, blocking all third party cookies may limit the functionality of some web resources, as some third party cookies may be used to enable some of their functions, for example related to web forms, databases, etc. Internet, with accounts in the client-bank system, etc.

[0126] Therefore, developers of the present technology have developed methods and systems for determining the probability value of a user, such as the first user 211 or the second user 213, acting on a third party web server 260 using its third party cookie. Specifically, in some embodiments, the present methods are designed to, for example, determine a probability value for the first user 211 to accept a third party cookie from a third party web server 260 when viewing a web page 302.

[0127] However, other non-limiting embodiments of the present technology are designed to determine the probability value of allowing the first user 211 to use a previously obtained version of a third party cookie in conjunction with a third party web server 260 when viewing a web page 302.

[0128] To determine such probability values, non-limiting embodiments of the present technology are designed to train an MLA algorithm, such as MLA algorithm 275, executed by training server 270, based on user-specific data of the first user 211, for example, containing the first user's past user actions. user 211 with a third party web resource 265. In addition, the present methods include performing, based on one of the probability values, a predetermined action, such as generating a notification advising the first user 211 to (1) allow or deny receiving a third party cookie, or (2) allow or deny cookie sharing -third party file with the third party web server 260, respectively, as described below.

[0129] Thus, the present methods and systems allow the determination of a user-specific probability value for a third party cookie, as described below. For example, if the first user 211 has frequently visited a third party web resource 265 in the past, the user may be advised to agree to receive a third party cookie from the third party web server 260 or to allow sharing of a third party cookie. Conversely, if the first user has never visited the third party's web resource 265, the user may be advised not to receive the third party's cookie or not be allowed to use it in conjunction with the third party's web server 260. It is contemplated that for other users, such as a second user 213 of a second electronic device 212, the MLA algorithm 275, trained as described below, may determine different values for the probability of being allowed to receive or share a third party cookie from the third party web server 260 sides.

[0130] It is contemplated that the training server 270 may be implemented similar to a first party web server 250 or a third party web server 260 and, therefore, may contain all or some of the elements of the computer system 100 described above with reference to FIG. 1.

[0131] It should be noted that there are no restrictions on the implementation of the MLA algorithm 275. In various non-limiting embodiments of the present technology, it may comprise an MLA algorithm implemented based on a neural network (NN), such as a convolutional NN (CNN, Convolutional NN) or a transformer-based NN, an MLA based inductive logic programming, support vector machine based MLA, clustering based MLA, Bayesian networks, etc. In some non-limiting embodiments of the present technology, the MLA algorithm 275 may be implemented as a decision tree MLA algorithm, such as a CatBoost decision tree MLA algorithm.

[0132] In some non-limiting embodiments of the present technology, the MLA algorithm 275 may be based on NN networks, such as a CNN network, a transformer-based NN network, and the like. In some non-limiting embodiments of the present technology, the MLA algorithm 275 may be implemented based on a decision tree model.

Communication network

[0133] In some non-limiting embodiments of the present technology, communications network 240 may be the Internet. In alternative, non-limiting embodiments of the present technology, the communications network 240 may be implemented as any suitable Local Area Network (LAN), Wide Area Network (WAN), private communications network, or the like. Obviously, the communication network embodiments are provided for illustrative purposes only. The implementation of the respective communication links (not specifically designated) between the first electronic device 210, the second electronic device 212, the first party web server 250, the third party server 260 and the training server 270 on the one hand and the communication network 240 on the other hand depends, among other things, from the implementation of the first electronic device 210, the second electronic device 212, the first party web server 250, the third party server 260, and the training server 270. By way of non-limiting example only, in those embodiments of the present technology where the first electronic device 210 is implemented in the form of a wireless communication device such as a smartphone, the communication link may be implemented as a wireless communication link. Examples of wireless communication links include a 3G network channel, a 4G network channel, etc. For example, communication network 240 may also utilize a wireless connection to a first party web server 250 and a first electronic device 210.

Machine learning model architecture

[0134] In FIG. 4 is a block diagram of a decision tree model 400 that can be used with at least some non-limiting embodiments of the present technology. Decision tree model 400 is intended to illustrate a general purpose decision tree model that can be modified to meet the requirements of a particular prediction model, such as MLA algorithm 275. For example, such modifications may include, but are not limited to, adding or removing one or more levels of the tree, adding or removing nodes (i.e., features and their associated branches), adding or removing branches connecting nodes and/or leaves tree.

[0135] The decision tree model 400 may be a trained tree model. In some embodiments of the invention, the generated decision tree model 400 may be updated and/or modified, for example, to increase the level of accuracy of the machine learning model and/or to expand the scope of the machine learning model. For example, in some embodiments, the decision tree model 400 may, among other things, be based on determining probability values for allowing or denying specific users, such as the first user 211 and the second user 213, to receive a third party cookie from the web server 260 third party while viewing the 302 web page, as described below. Other applications of the decision tree model 400 are also possible without going beyond the scope of the present technology.

[0136] Decision tree model 400 includes a first node 402 associated with the first feature f1. The first node 402 forms the first level of the decision tree model 400. The first node 402 is connected by branches to the second node 404 and to the third node 406. The second node 404 and the third node 406 are connected to the second feature f2. The second node 404 and the third node 406 form the second level of the decision tree model 400. In some embodiments, a first feature f1 (and a branch for the first feature f1) is selected from a set of features to be placed at the first level of a decision tree model 400 based on a plurality of training digital objects. For example, this feature may indicate past user activity of training users on various first-party and third-party web properties, as described below.

[0137] The first feature f1 is selected such that, for a digital object, the value of the parameter associated with the first feature f1 determines whether the digital object should be associated with the second node 404 or the third node 406. For example, if this value is less than the value of f1, then the digital object is associated with the second node 404. In another example, if this value is greater than the value of f1, then the digital object is associated with the third node 406.

[0138] In turn, the second node 404 is connected to a fourth node 408 associated with the third feature f3, and to a fifth node 410 associated with the third feature f3. The third node 406 is connected to a sixth node 412 associated with a third feature f3 and to a seventh node 414 associated with a third feature f3. A fourth node 408, a fifth node 410, a sixth node 412, and a seventh node 414 form the third level of the decision tree model 400. As described above with respect to the first node 402, for a digital object, the value of the parameter associated with the second attribute f2 determines whether the digital object should be associated with the fourth node 408 or the fifth node 410 (if the digital object is associated with the second node 404) or the sixth node 412 or the seventh node 414 (if the digital object is associated with the third node 406).

[0139] In turn, all nodes among the fourth node 408, fifth node 410, sixth node 412, and seventh node 414 are associated with sets of predicted parameters. In the one shown in FIG. 4, the predicted parameter sets include a first set 420, a second set 422, a third set 424, and a fourth set 426. Each predicted parameter set contains three target values C1, C2, and C3, for example, provided by training users, as described below.

[0140] As will be appreciated by one skilled in the art, decision tree model 400 illustrates an embodiment of the invention in which a particular level of decision tree model 400 is associated with a single feature. In the one shown in FIG. 4 example, the first level contains the first node 402 and is associated with the first feature f1, the second level contains the second node 404 and the third node 406 and is associated with the second feature f2, the third level contains the fourth node 408, the fifth node 410, the sixth node 412 and the seventh node 414 and is associated with the third feature f3.

[0141] In other words, as shown in FIG. In embodiment 4, the first level is associated with the first feature f1, the second level is associated with the second feature f2, the third level is associated with the third feature f3. In this case, other embodiments of the invention are also possible. In particular, an alternative embodiment of the invention is possible in which the generated tree model may contain various features for some level of the tree model. For example, the first level of such a tree model may contain a first node associated with the first feature f1, the second level may contain a second node associated with the second feature f2, and a third node associated with the third feature f3. As will be clear to one skilled in the art of the present technology, other options for matching features to a level are possible without going beyond the boundaries of the present technology.

[0142] There is no restriction on the selection of features such as the first feature f1, the second feature f2, and the third feature f3 to form the decision tree model 400. In some non-limiting embodiments of the present technology, selecting features, and therefore branches between nodes of the decision tree model 400, may include optimizing a metric indicative of the quality of the feature for separating into branches of the decision tree model 400 a plurality of training digital objects during the training process. as each object from the plurality of training digital objects is fed into the decision tree model 400 . In some non-limiting embodiments of the present technology, such a metric may include an Estimate of Positive Correctness metric, a Gini Impurity metric, an Information Gain metric, and the like. In specific non-limiting embodiments of the present technology, features may be selected based on maximizing the prediction quality parameter of the proto-trees that comprise the decision tree model 400, for example, as described in patent application US2019164084A1 "Method of and system for generating prediction quality parameter for a prediction model executed in a machine learning algorithm" (published May 30, 2019), the contents of which are incorporated herein by reference in their entirety.

[0143] Additionally, according to some non-limiting embodiments of the present technology, MLA algorithm 275 may comprise one instance of decision tree model 400. In this case, the MLA algorithm can be called a "single decision tree". In other non-limiting embodiments of the present technology, the MLA algorithm 275 may comprise an ensemble of decision trees implemented similar to the decision tree model 400, in which case the MLA algorithm 275 may be referred to as a “decision tree forest.” In this case, the final prediction of the ensemble of decision trees is based on the prediction results of each decision tree in the ensemble. There are no restrictions on constructing an ensemble of decision trees; for example, it may involve using a boosting approach such as a gradient boosting approach.

[0144] In general, boosting is a method of constructing an ensemble of decision trees, where the trees are formed in stages (for example, as opposed to building a random forest, where trees can be formed in parallel) such that the difference between the target values and the predicted outputs of the ensemble of decision trees is minimized at each iteration when forming each new decision tree. For example, the difference between target values and prediction results for an ensemble of decision trees can be expressed by a loss function, such as a cross-entropy loss function.

[0145] It should be understood that in non-limiting embodiments of the present technology, other embodiments of the loss function are also possible, such as the root mean square error loss function, the Huber loss function, the piecewise linear loss function, etc.

[0146] In other words, using a boosting approach, the MLA algorithm 275 uses decision trees as weak learners, which the MLA algorithm 275 sequentially combines so that each new learner matches the residuals from the previous iteration, thereby improving the aggregate result of the entire ensemble of trees decisions. In other words, each decision tree is built on the same set of training digital objects, while the training digital objects for which the first decision tree made “errors” in forecasting have priority when constructing the second decision tree, etc. The 275 MLA algorithm then combines the results of each iteration to form a strong learner.

[0147] In specific non-limiting embodiments of the present technology, the MLA algorithm 275 may comprise an ensemble of CatBoost decision trees, such as described in "CatBoost: gradient boosting with categorical features support" (Dorogush et al., published by Yandex Inc. https://arxiv.org/abs/1706.09516), the contents of which are incorporated herein by reference in their entirety.

[0148] In general, it can be said that the processor of the training server 270 and/or the first electronic device 210 must perform two corresponding processes related to the MLA algorithm 275. The first process of these two processes is a training process, where the processor 110 of the training server 270 is capable of training the MLA algorithm 275, based on a set of training digital objects (also called a training data set), to determine the probability value of the first user 211 to agree to receive a third party cookie. from the third party web server 260 when viewing the web page 302. The learning process is described in detail below with reference to FIG. 5.

[0149] The second process is a use stage process where the processor 110, for example, of the first electronic device 210, executes the MLA algorithm 275 thus trained to determine such a probability value, and then based on it can cause the browser application 215 to generate an appropriate advisory notification for the first user 211. The process of the use step is described in detail below with reference to FIG. 6 and 7.

[0150] According to non-limiting embodiments of the present technology, the training process and the use phase process may be performed by the training server 270 and/or the processor 110 of the first electronic device 210 in the network computing environment 200 described above.

Learning process

[0151] According to some non-limiting embodiments of the present technology, the training server 270 can train the MLA algorithm 275 based on a plurality of training digital objects generated from data about the past user activities of various training users with various training web pages containing related them third party web resources. In fig. 5 illustrates a step diagram for generating a learning digital object 512 from a plurality of learning digital objects according to some non-limiting embodiments of the present technology.

[0152] As shown in FIG. 5, in some non-limiting embodiments of the present technology, to generate a learning digital object 512, the learning server 270 may access the learning electronic device 510 of a learning user 511 from a plurality of learning users 509. In some non-limiting embodiments of the present technology, a plurality of learning users 509 may be general users of the communications network 240, such as the first user 211 and the second user 213. In other non-limiting embodiments of the present technology, the plurality of training users 509 may be human evaluators who have been explicitly instructed to visit training web pages and perform specific actions with them, such as allowing or denying third party cookies from them.

[0153] It is apparent that the teaching electronic device 510 may be implemented similar to the first electronic device 210 and the second electronic device 212 described above. In addition, it is contemplated that the educational electronic device 510 may also execute a browser application implemented similar to the browser application 215 described above, thereby allowing the educational user 511 to view various web resources available over the communication network 240.

[0154] Thus, according to some non-limiting embodiments of the present technology, in order to generate a learning digital object 512, the learning server 270 may: (a) access the learning electronic device 510 through a communications network 240 and (b) receive data from it about past user activities of the training user 511 with various training web pages, such as the training web page 502. It is contemplated that the training electronic device 510 can provide data about the training user 511 to the training server 270 without going beyond the boundaries of the present technology.

[0155] Similar to the above, in some non-limiting embodiments of the present technology, the educational web page 502 may contain specific third party elements that are associated with the third party educational web resource 504 and link the third party educational web resource 504 with the educational web page 502. It should be understood that there are no restrictions on the implementation of the training web page 502. In some non-limiting embodiments of the present technology, it may be implemented like a web page 302. The third party educational web resource 504 may also be implemented like a third party web resource 265, as described above with reference to FIG. 3.

[0156] Thus, in some non-limiting embodiments of the present technology, training server 270 may generate a training digital object 512 containing (a) third party training web resource-dependent data of a third party training web resource 504 associated with the training web page 502; (b) training user-specific data from training user 511 viewing training web page 502; and (c) a label 505 indicating whether or not the training user 511 is allowed to receive a third party training cookie from the third party training web resource 504 when viewing the training web page 502.

[0157] For example, according to some non-limiting embodiments of the present technology, the third party training web resource-dependent data of the third party training web resource 504 may comprise the domain name of the third party training web resource 504, the domain zone of the training web resource Third party 504 (such as *.com, *net, *.org or country code such as *.ca, *co.uk), third party 504 web property category, etc. For example, as described above, the following categories of third party educational web resource 504 may be used: (a) social network and/or (b) news portal and/or (c) video hosting platform. Other categories of third party educational web resource 504 are also possible. It should be understood that the data dependent on the third party web resource 504 may further comprise other data of the third party web resource 504, such as the name of the registrar of the third party web resource 504, the name of the hosting service provider for the teacher third party web resource 504 or specific web statistics of a third party training web resource 504, without going beyond the boundaries of the present technology.

[0158] In addition, according to some non-limiting embodiments of the present technology, the training user-specific data of the training user 511 may contain various data about the past user activities of the training user 511 with the third party training web resource 504, which, for example, may contain (a) the total number of visits by the training user 511 to the third party training web resource 504 and (b) the number of visits by the training user 511 to the third party training web resource 504 during a predetermined period, such as a day, a week, a month, etc. . Other past user activities of the training user 511 with the third party training web resource 504 that may be considered part of the training user-dependent data may include (a) the average time spent by the training user 511 on the third party training web resource 504 in during the day, week, month, etc.; (b) the number of “clicks” on the web pages of the third party educational web resource 504 during the day, week, month, etc.; (c) the number of scrolls on the web pages of the third party educational web resource 504, etc.

[0159] Additionally, in some non-limiting embodiments of the present technology, the training user-specific data of the training user 511 may further include data about the training user's past user activities with the training web page 502, which may contain data similar to that described above for third party web training resource.

[0160] In some non-limiting embodiments of the present technology, the training user-specific data may further comprise user profile data of the training user 511, which, for example, includes, among other things, the age of the training user 511, the gender of the training user 511, the user's socio-economic a parameter of the training user 511, such as his level of education, specialty, estimated average income, etc.

[0161] Additionally, in some non-limiting embodiments of the present technology, the training user-specific data of the training user 511 may further comprise data of the training electronic device 510, which, for example, includes, among other things: the make and model of the training electronic device 510; the operating system of the educational electronic device 510 and its version; data from applications installed on the educational electronic device 510; data of web pages visited from the educational electronic device 510 during a predetermined period, such as a day, a week, a month, etc.; browser application (or multiple application) data such as view and version; data from sensors of the educational electronic device 510, such as its GPS sensor, allowing, for example, to determine the current geographic location of the educational user 511. Other data from the educational electronic device 510 can also be used to determine educational user-specific data without going beyond the boundaries of the present technology.

[0162] Additionally, according to some non-limiting embodiments of the present technology, the associated label 505 may be binary values such as "1" or "0", "Yes" or "No". "), "Accepted", or "Declined", indicating that the training user 511 has allowed or denied receiving a third party cookie associated with the third party training web resource 504.

[0163] Additionally, in some non-limiting embodiments of the present technology, the training server 270 may assign a label 505 only if the training user 511 has explicitly allowed or disabled the receipt of a training cookie from a third party training web resource 504 when viewing a training web page 502, for example, (1) in response to a corresponding notification generated by the browser application of the educational electronic device 510, including one similar to notification 310 (as described above), or (2) using the browser application's preference settings. Specifically, in these embodiments, the learning server 270 may ignore those learning digital objects where third party learning cookies are accepted or rejected by the default settings of the learning electronic devices' browser applications. However, in other non-limiting embodiments of the present technology, the learning server 270 may consider any options to allow or disable the receipt of a third party cookie from a third party web resource 504 when generating the learning digital object 512, even if the learning browser application electronic device 510 allows or denies receipt of a third party training cookie when training user 511 is unaware of it.

[0164] Thus, the training server 270 can generate a plurality of training digital objects based on each user's past user actions from the plurality of training users 509 with various training web pages and associated third party training web resources (as described above). In addition, the training server 270 can generate for each object from the plurality of training digital objects a corresponding feature vector, which the training server 270 can also use to train the MLA algorithm 275.

[0165] Specifically, according to some non-limiting embodiments of the present technology, to train MLA algorithm 275, training server 270 may sequentially move each object of a plurality of training digital objects down branches of each tree of an ensemble of decision trees implemented as described above with reference to FIG. 4. Specifically, by downward movement of training digital object 512, training server 270 can determine branches between tree nodes from an ensemble of decision trees based on features of the training feature vector. Using, for example, a boosting approach to form an ensemble of decision trees, as described above with reference to FIG. 4, training server 270 may minimize a loss function corresponding to the differences between the output of the ensemble at its current iteration of its formation and its target values, represented by labels assigned to each object of the plurality of training digital objects, as described above. Thus, training server 270 can expand the ensemble (or forest) of decision trees until a predetermined difference threshold for the loss function is reached.

[0166] The training server 270 may train the MLA algorithm 275, thereby determining the MLA algorithm formula 285 for the probability value of allowing or denying a user of the communications network 240, such as the first user 211 or the second user 213, to receive a third party cookie from the web. third party server 260 when viewing web page 302 (see FIG. 2).

[0167] However, as described above, in other non-limiting embodiments of the present technology, the training server 270 may train the MLA algorithm 275 to determine the probability value of allowing the first user 211 to use a third party cookie previously received by the first electronic device 210, in conjunction with a third party web server 260. In these embodiments, training server 270 may prepare training data in a different manner. In particular, instead of the plurality of learning digital objects described above, the learning server 270 may generate, based on data about past user actions of learning users such as learning user 511, a second set of learning digital objects, each of which differs from the learning digital object 512 only in structure assigned label.

[0168] In particular, the label assigned to the learning digital object from the second set of learning digital objects indicates whether the learning user 511 is allowed or prohibited from using a third party learning cookie previously received by the learning electronic device 510 from the third party learning web resource 504 party, in conjunction with its third party training web server when viewing the training web page 502.

[0169] The training server 270 may train the MLA algorithm 275 based on the second set of training digital objects in a manner similar to that described above. Accordingly, as a result of training the MLA algorithm 275, the training server 270 may determine a second MLA algorithm formula for the probability value of allowing or denying a user of the communication network 240, such as the first user 211 or the second user 213, to use a third party cookie previously received from the web. - a third party server 260, in conjunction with this server when viewing the web page 302.

[0170] The training server 270 may send the MLA algorithm formula 285 to the corresponding electronic devices connected to the communication network 240 to perform the use step process described below with reference to FIG. 6 and 7.

Use phase process

[0171] According to some non-limiting embodiments of the present technology, when performing the use step process, the first electronic device 210 and/or the second electronic device 212 may obtain the MLA algorithm formula 285 for further use to determine the probability value of the first user to allow or deny 211 or a second user 213 to receive a third party cookie from the third party web server 260 when viewing the web page 302 (see FIGS. 2 and 3). In particular, according to some non-limiting embodiments of the present technology, the first electronic device 210 may (a) obtain first use phase data; (b) generate the first vector of features of the use stage; (c) apply, with respect to the first use phase feature vector, the MLA algorithm formula 285 to determine a first value for the probability of allowing or denying the first user 211 to receive a third party cookie; and (d) perform a predetermined action based on the first probability value.

[0172] It is contemplated that, according to some non-limiting embodiments of the present technology, the first use phase data may be similar to the training data and may comprise (a) first user dependent data of the first user 211, similar to the training user dependent training data user 511 (as described above), and/or (b) usage phase third party web resource dependent data of the third party web resource 265 hosted on the third party web server 260, similar to the third party training web resource dependent data third party educational web resource 504 (as described above).

[0173] In some non-limiting embodiments of the present technology, after determining a first probability value associated with the first user 211, the first electronic device 210 may compare the first probability value to a predetermined threshold probability value, which, for example, may be 0. 75, 0.80 or 0.85, and based on the result of this comparison, perform a predetermined action. For example, if the first probability value is greater than or equal to a predetermined probability threshold, then the first electronic device 210 may determine that the first user 211 is likely to allow the receipt of a third party cookie from the third party web server 260.

[0174] Accordingly, in some non-limiting embodiments of the present technology, the first electronic device 210 may further cause the browser application 215 to generate a first notification 602 (see FIG. 6). It is contemplated that the first notification 602 may notify the first user 211 that it can allow the third party cookie to be received from the third party server 260. In alternative, non-limiting embodiments of the present technology, if the first probability value is greater than or equal to a probability threshold, then the first electronic device 210 may cause the browser application 215 to automatically allow the third party cookie to be received from the third party server 260 without any or notifications.

[0175] However, if the first probability value is less than a probability threshold, then the first electronic device 210 may determine that it is unlikely that the first user 211 will allow a third party cookie from a third party web resource 265 when viewing the web page 302. Or the first the electronic device 210 may determine that the first user 211 is likely to deny receiving a third party cookie.

[0176] Accordingly, in some non-limiting embodiments of the present technology, the first electronic device 210 may cause the browser application 215 to generate a second notification 702 (see FIG. 7). It is contemplated that the second notification 702 may advise the first user 211 not to allow a third party cookie to be received from the third party web server 260. In alternative non-limiting embodiments of the present technology, if the first probability value is less than a probability threshold, then the first electronic device 210 may cause the browser application 215 to automatically prevent the browser application 215 from receiving a third party cookie from the third party server 260 without any notification.

[0177] In other non-limiting embodiments of the present technology, the first electronic device 210 may cause the browser application 215 to perform a predetermined action only in response to determining that the first user 211 is unlikely to allow a third party cookie. That is, in these embodiments, (a) if the first probability value is greater than or equal to the probability threshold, then the first electronic device 210 may trigger the browser application 215 to automatically allow the third party cookie without any notification, and ( b) if the first probability value is less than the probability threshold, then the first electronic device 210 may cause the browser application 215 to generate a second notification 702 advising the first user 211 not to allow the receipt of a third party cookie from the third party web server 260.

[0178] In other non-limiting embodiments of the present technology, the first electronic device 210 may otherwise perform predetermined actions based on the first probability value. For example, once a first probability value is determined, the first electronic device 210 may compare it to various probability thresholds, such as a lower predetermined threshold value, an average predetermined threshold value, and an upper predetermined threshold value, each of which is greater than the previous value. For example, the lower predetermined threshold value may be 0.75, the average predetermined threshold value may be 0.85, and the upper predetermined threshold value may be 0.95. Thus, in some non-limiting embodiments of the present technology, if the first probability value is between the middle and upper predetermined threshold values, then the first electronic device 210 may cause the browser application 215 to generate a first notification 602. If the first probability value is greater than the upper predetermined threshold a predetermined threshold value, then the first electronic device 210 may trigger automatic permission of the browser application 215 to receive a third party cookie from the third party web server 260 without any notification.

[0179] In some non-limiting embodiments of the present technology, if the first probability value is between the middle and lower predetermined threshold values, then the first electronic device 210 may cause the browser application 215 to generate a second notification 702. If the first probability value is less than the lower predetermined threshold a predetermined threshold value, then the first electronic device 210 may cause the browser application 215 to automatically prevent the browser application 215 from receiving a third party cookie from the third party web server 260 without any notification.

[0180] Just as the first electronic device 210 can generate a first probability value associated with the first user 211, in some non-limiting embodiments of the present technology, the second electronic device 212 is capable of using the MLA algorithm formula 285 to determine a second probability value for resolving the second user 213 to receive a third party cookie from third party web server 260 when viewing web page 302 (see FIG. 2). To this end, the second electronic device 212 may receive second use-phase data comprising (a) second-user-dependent data of the second user 213 similar to the first-user-dependent data of the first user 211 (as described above), and/or (b) dependent from the third party web resource of the use stage, data from the third party web resource 265 hosted on the third party web server 260.

[0181] Thus, since the second user-dependent data of the second user is generally different from the first user-dependent data of the first user 211, the second electronic device 212 can, using the MLA algorithm formula 285, determine a second probability value different from the first probability value. Additionally, based on the second probability value, the second electronic device 212 may perform a predetermined action similar to that described above for the first electronic device 210, such as generating a first or second notification 602, 702, or automatically allowing or denying the receipt of a third cookie. side from a third party web server 260.

[0182] Thus, the present systems and methods are capable of providing a highly individualized recommendation and/or decision regarding whether to allow or disallow receiving a third party cookie from a third party web server 260 for each user of the communications network 240, such as the first user 211 and the second user 213.

[0183] Additionally, in those embodiments where a third party cookie has previously been received from a third party web server 260, for example, by the first electronic device 210, the first electronic device 210 may obtain the second MLA algorithm formula from the training server 270. The first electronic device 210 may apply the second MLA formula to the first use phase feature vector to determine a third probability value of allowing the first user 211 to use the third party cookie in conjunction with the third party web server 260. Likewise, the first electronic device 210 may further compare the third probability value with a second predetermined probability threshold value and perform a predetermined action based on the results of this comparison. For example, as with the first probability value, the first electronic device 210, based on the third probability value, may cause the browser application 215 to generate an appropriate type of first or second notification 602, 702 advising the first user 211 to allow or disable the use of a third party cookie in conjunction with third party web server 260, respectively.

[0184] However, it should be understood that other predetermined actions are also possible, such as automatically sharing or automatically preventing sharing of a third party cookie by the first electronic device 210. It should also be noted that the first electronic device 210 may perform such predetermined actions in a manner similar to that described in detail above for embodiments of the invention designed to determine a first probability value of allowing the first user 211 to receive a third party cookie.

[0185] Obviously, similar to determining the second probability value, the second electronic device 212, using the second formula of the MLA algorithm, can determine a fourth probability value of allowing the second user 213 to use the third party cookie with the third party server 260, which is different from the third value probability associated with the first user 211.

[0186] Moreover, according to some non-limiting embodiments of the present technology, after performing a predetermined action as described above, the first electronic device 210 may (1) observe further user actions of the first user 211 in response to the provision of the first notification 602 or second notification 702 and (2) send an indication of a further user action to the training server 270 to update the training data set used to train the MLA algorithm 275 if the further user action does not correspond to the first notification 602 or the second notification 702. For example, in some In limiting embodiments of the present technology, if it is determined that, contrary to the first notification 602 (see FIG. 6), the first user 211 has rejected a third party cookie, then the first electronic device 210 may generate an additional educational digital object containing (a) dependent the first user of the use stage, the first user data 211; (b) third party web resource-dependent usage phase data of the third party web resource 265; and (c) a usage milestone indicating that the first user 211 has rejected the third party cookie from the third party web server 260. Additionally, in some non-limiting embodiments of the present technology, the first electronic device 210 may send, via communications network 240, an additional training digital object to the training server 270 to update the plurality of training digital objects used to train the MLA algorithm 275, as described above. Thus, the training server 270 can periodically update the MLA algorithm formula 285 to improve its accuracy in determining probability values for users of the communications network 240.

[0187] Although the above-described process of the use step is performed by the first and second electronic devices 210, 212 using the MLA algorithm formula 285 or the second MLA algorithm formula sent to them by the training server 270, it should be understood that in other non-restrictive In nature, in embodiments of the present technology, the use phase process may also be performed by the training server 270 based on the first and second use phase data sent to it by the first and second electronic devices 210, 212, respectively. For example, in these embodiments, in response to receiving the first or second use phase data, the training server 270 may further determine a corresponding first or second probability value to send back to the first and second electronic devices 210, 212, respectively.

First way

[0188] The architecture and examples described above enable a method of training an MLA algorithm, such as MLA algorithm 275, to determine probability values for allowing target users of the communications network 240, such as the first user 211 or the second user 213, to receive third party cookies when browsing 302 web pages as described above. In fig. 8 is a flow diagram of a first method 800 according to non-limiting embodiments of the present technology. The first method 800 may be performed by the processor 110 of the training server 270.

Step 802: The server receives a plurality of training objects from the plurality of training electronic devices.

[0189] The first method 800 begins at step 802, in which the learning server 270 may receive a plurality of learning digital objects from a plurality of learning electronic devices associated therewith, as described above with reference to FIG. 5. As mentioned above, in some non-limiting embodiments of the present technology, the training server 270 can obtain a plurality of training digital objects by accessing electronic devices connected to the communication network 240. However, in other non-limiting embodiments of the present technology, educational electronic devices may be associated with a plurality of training users 509, which are human evaluators who are explicitly instructed to visit various training web pages and determine whether third party cookies are required. accepted or rejected.

[0190] In particular, as described above with reference to FIG. 5, in some non-limiting embodiments of the present technology, an educational digital object 512 of a plurality of educational digital objects may comprise (a) third party educational web resource-dependent data of a third party educational web resource 504 associated with an educational web page 502 ; (b) training user-specific data from training user 511 viewing training web page 502; and (c) a label 505 indicating whether or not the training user 511 is allowed to receive a third party training cookie from the third party training web resource 504 when viewing the training web page 502.

[0191] In those embodiments of the present technology where training server 270 can train MLA algorithm 275 to determine the probability value of first user 211 allowing a third party cookie previously received by first electronic device 210 in conjunction with third party web server 260, training server 270 may otherwise prepare training data. In particular, instead of the plurality of learning digital objects described above, the learning server 270 may generate, based on data about past user actions of learning users such as learning user 511, a second set of learning digital objects, each of which differs from the learning digital object 512 only in structure the label assigned to it.

[0192] In particular, the label assigned to the learning digital object from the second plurality of learning digital objects indicates whether the learning user 511 is allowed or prohibited from using a third party learning cookie previously received by the learning electronic device 510 from the third party learning web resource 504 party, in conjunction with its third party training web server when viewing the training web page 502.

[0193] The first method 800 then continues at step 804.

Step 804: The server inputs a plurality of training objects into the MLA algorithm and thereby determines the MLA algorithm formula for the probability values of target users allowing a third party cookie from the third party web resource of the usage step.

[0194] At step 804, the training server 270 may input the resulting set of training digital objects into the MLA algorithm 275, thereby training the MLA algorithm 275 to determine the probability values of allowing the first user 211 and the second user 213 to receive a third party cookie from the web server 260 third party when viewing a 302 web page.

[0195] To this end, in those non-limiting embodiments of the present technology where the MLA algorithm 275 is an ensemble of decision trees, as described above with reference to FIG. 4, training server 270 may sequentially move each object of the plurality of training digital objects down branches of each decision tree ensemble tree, thereby determining branches between nodes of the decision tree ensemble tree.

[0196] Using, for example, a boosting approach to form an ensemble of decision trees, as described above with reference to FIG. 4, training server 270 may minimize a loss function corresponding to the differences between the output of the ensemble at its current iteration of its formation and its target values, represented by labels assigned to each object of the plurality of training digital objects, as described above.

[0197] At this time, the training server 270 may train the MLA algorithm 275, thereby defining the MLA algorithm formula 285 for the probability values of allowing the first user 211 and the second user 213 to receive a third party cookie from the third party web server 260 when browsing the web. pages 302.

[0198] Accordingly, based on the second set of training digital objects, the training server 270 may train the MLA algorithm 275 to determine a second MLA algorithm formula for the probability value of allowing or denying a user of the communications network 240, such as the first user 211 or the second user 213, to use a cookie. a third party file previously received from a third party web server 260, shared with that server when viewing the web page 302.

[0199] The first method then proceeds to step 806.

Step 806: The server sends an MLA algorithm formula to the at least one target electronic device of the target user to determine a probability value of allowing the target user to receive a third party cookie from the third party web resource of the use step.

[0200] At step 806, the training server 270 may send the MLA algorithm formula 285 and/or the second MLA algorithm formula to the first electronic device 210 and the second electronic device 212 for subsequent use in determining the probability values described above.

[0201] This completes the first method 800.

Second way

[0202] The architecture and examples described above enable a method for managing tracking of web browsing operations of a target user in a browser application, for example, a first user 211 in a browser application 215 performed by a first electronic device 210 as described above. In fig. 9 is a flow diagram of a second method 900 according to non-limiting embodiments of the present technology. The second method 900 may, for example, be performed by the processor 110 of the first electronic device 210.

Step 902: The electronic device obtains from the web server data representing the use stage web page for display in the browser application.

[0203] The second method 900 begins at step 902, in which the first electronic device 210 may, for example, in response to the first user 211 submitting a corresponding URL in the address bar of the browser application 215, obtain executable source code from the first party web server 250 web page 302 of web resource 255. The first electronic device 210 may then cause the browser application 215 to render a rendering of the web page 302 for presentation to the first user 211, as schematically shown in FIG. 3.

[0204] As described above, examples of web resource 255 containing web page 302 include a website such as a commercial website designed to advertise specific products, a news portal (such as RIA Novosti™ or CNN ™), online video service (for example, Kinopoisk.ru™), etc.

[0205] The second method 900 then continues at step 904.

Step 904: The electronic device determines, based on the data representing the use stage web page, elements thereof associated with at least one third party web resource of the use stage.

[0206] According to some non-limiting embodiments of the present technology, as described above with reference to FIG. 3, at step 904, the first electronic device 210 may, among the plurality of layout elements making up the web page 302, determine third party elements 308, such as third party element 309. As described above, the third party layout element 309 may represent the content of a third party web resource 265 hosted on a third party web server 260 different from the first party web server 250 containing the web resource 255. Thus, the web resource 265 the third party may be linked to the web page 302 through the third party element 309. As described above, as used herein, the term “linked” means that the third party element 309 may (a) redirect the user 211 viewing the web page 302 to a corresponding web page of the third party web resource 265 in response to a predetermined action the first user 211 with a third party element 309, such as "clicking" on it, and/or (b) send to the third party web server 260 specific data indicating actions on only the third party element 309 and/or on the entire web page 302.

[0207] Additionally, according to some non-limiting embodiments of the present technology, web resource 255 may use cookies, including third party cookies, such as a third party cookie from third party web server 260 that may be sent to the first electronic device 210. Thus, by sending a third party cookie to the first electronic device 210, the third party web server 260 can track the web browsing activities of the first user 211, such as, but not limited to, (a) browsing history web pages by the first user 211; (b) clicking on buttons or following links on various web pages; (c) search operations performed on search engines; (d) starting or ending a session on various resources, such as web email services, social networks, etc. In additional non-limiting embodiments of the present technology, the data that may be made available to the third party web server 260 using a third party cookie may further include data from the first electronic device 210, including, but not limited to, (a) the model of the first electronic device 210; (b) information about the operating system, such as versions, dates of last updates, etc.; (c) data from sensors of the first electronic device 210, such as GPS data; etc.

[0208] Because there may be risks to the personal information of the first user 211 as a result of sharing such data with third party web servers, the second method 900 proceeds to the next steps to determine whether the first user 211 should block or allow the third party to receive a cookie from 260 third party web server.

[0209] The second method 900 then continues at step 906.

Step 906: The electronic device receives the usage phase data.

[0210] At step 906, the first electronic device 210 may receive the first use phase data comprising the first user-dependent data of the first user 211 and/or the third-party web resource-dependent data of the third party web resource 265 hosted on the web server 260 a third party, as described above for the process of the use step of the MLA algorithm 275.

[0211] The second method 900 then continues at step 908.

Step 908: The electronic device inputs the usage phase data into the MLA algorithm to determine a probability value of the target user allowing the at least one third party web resource to receive a usage phase third party cookie while viewing a usage phase web page, where A third party usage cookie enables the third party server to track the target user's web browsing activity within the browser application.

[0212] According to some non-limiting embodiments of the present technology, at step 908, the first electronic device 210 may apply to the first use phase data obtained in the previous step an MLA algorithm 275 trained in accordance with the first method 800 as described above. . To this end, in some non-limiting embodiments of the present technology, the first electronic device 210 may receive from the training server 270 an MLA algorithm formula 285 generated by the MLA algorithm 275 as a result of executing the first method 800.

[0213] The first electronic device 210 may then apply the MLA algorithm formula 285 to the first usage phase data to generate a first probability value of allowing the first user 211 to receive a third party cookie from the third party web server 260 when viewing the web page 302 As described above, by allowing the first user 211 to receive a third party cookie, the third party web server 260 can track the web browsing activities of the first user 211.

[0214] However, in other non-limiting embodiments of the present technology, to determine a first probability value associated with the first user 211, the first electronic device 210 may send first usage phase data via the communication network 240 to the training server 270. In response, the training server 270 may apply the MLA algorithm formula 285 to the first use phase data to determine a first probability value and send the first probability value back to the first electronic device 210.

[0215] As described above, the second electronic device 212 may also receive from the training server 270 an MLA algorithm formula 285 to determine, based on the second usage phase data, a second probability value of allowing the second user 213 to receive a third party cookie from the web server 260 third party when viewing web page 302. It is assumed that since this determination is made based on other usage stage data, the second probability value may be different from the first probability value.

[0216] The second method 900 then proceeds to step 910.

Step 910: The electronic device determines that it is unlikely that the target user will allow the user to receive a third party cookie of the usage stage from at least one third party web resource of the usage stage while browsing the web page of the usage stage if the probability value is less than a probability threshold; and generating a first notification by the electronic device advising the target user to opt-out of receiving a third-party usage-stage cookie from the at least one third-party usage-stage web property.

[0217] At step 910, the first electronic device 210 may, based on the first probability value, (a) determine that the first user 211 is likely to accept or reject a third party cookie from the third party web server 260, and (b) perform in advance specified action.

[0218] For example, if it is determined that the first probability value is greater than or equal to a predetermined probability threshold, then the first electronic device 210 may determine that the first user 211 is likely to accept a third party cookie from the third party web server 260 .

[0219] Accordingly, in some non-limiting embodiments of the present technology, the first electronic device 210 may cause the browser application 215 to generate a first notification 602 advising that there is no threat to the first user 211's confidential data if the third user is allowed to receive a cookie. side from the third party server 260 (as described above with reference to FIG. 6). In other non-limiting embodiments of the present technology, if the first probability value is greater than or equal to the probability threshold, then the first electronic device 210 may cause the browser application 215 to automatically allow the third party cookie to be received from the third party server 260 without any or notifications.

[0220] Conversely, if the first probability value is determined to be less than a probability threshold, then the first electronic device 210 may determine that it is unlikely that the first user 211 will allow a third party cookie from a third party web resource 265 when viewing the web page 302 Or, the first electronic device 210 may determine that the first user 211 is likely to deny receiving a third party cookie.

[0221] Accordingly, in some non-limiting embodiments of the present technology, the first electronic device 210 may cause the browser application 215 to generate a second notification 702 advising the first user 211 not to receive a third party cookie from the third party web resource 265 (as described above with reference to Fig. 7).

[0222] In some non-limiting embodiments of the present technology, the first electronic device 210 may cause the browser application 215 to perform a predetermined action only in response to determining that the first user 211 is unlikely to allow a third party cookie. That is, in these embodiments, (a) if the first probability value is greater than or equal to the probability threshold, then the first electronic device 210 may trigger the browser application 215 to automatically allow the third party cookie without any notification; (b) if the first probability value is less than the probability threshold, then the first electronic device 210 may cause the browser application 215 to generate a second notification 702 advising the first user 211 not to allow the receipt of a third party cookie from the third party web server 260.

[0223] Moreover, according to some non-limiting embodiments of the present technology, after performing a predetermined action as described above, the first electronic device 210 may (1) observe further user actions of the first user 211 in response to the provision of the first notification 602 or second notification 702 and (2) send an indication of a further user action to the training server 270 to update the training data set used to train the MLA algorithm 275 if the further user action does not correspond to the first notification 602 or the second notification 702. In particular, in some In limited embodiments of the present technology, if it is determined that, contrary to the first notification 602 (as described above with reference to FIG. 6), the first user 211 has rejected a third party cookie, then the first electronic device 210 may generate an additional learning digital object containing (a) the first user-dependent data of the first user 211; (b) third party web resource-dependent usage phase data of the third party web resource 265; and (c) a usage milestone indicating that the first user 211 has rejected the third party cookie from the third party web server 260. Additionally, in some non-limiting embodiments of the present technology, the first electronic device 210 may send, via communications network 240, an additional training digital object to the training server 270 to update the plurality of training digital objects used to train the MLA algorithm 275, as described above. Thus, the training server 270 can periodically update the MLA algorithm formula 285 to improve its accuracy in determining probability values for users of the communications network 240.

[0224] Additionally, in some non-limiting embodiments of the present technology, the first electronic device 210 and the second electronic device 212 may obtain a second MLA formula from the training server 270. For example, the first electronic device may apply a second MLA formula to the first use phase feature vector to determine a third probability value of allowing the first user 211 to use a third party cookie previously received by the first electronic device 210 in conjunction with the third party web server 260. Likewise, the first electronic device 210 may further compare the third probability value with a second predetermined probability threshold value and perform a predetermined action based on the results of this comparison. For example, the first electronic device 210, based on the third probability value, may cause the browser application 215 to generate the appropriate type of first or second notifications 602, 702 that advise the first user 211 to allow or disallow the use of a third party cookie in conjunction with the third party web server 260. respectively.

[0225] It should be understood that other predetermined actions are possible, such as automatically sharing or automatically disabling the sharing of a third party cookie by the first electronic device 210. The first electronic device 210 may perform such predetermined actions in a manner similar to as described in detail above for embodiments of the invention designed to determine a first probability value of allowing the first user 211 to receive a third party cookie.

[0226] Obviously, similar to determining the second probability value, the second electronic device, using the second formula of the MLA algorithm, can determine a fourth probability value of allowing the second user 213 to use the third party cookie with the third party server 260, which is different from the third probability value associated with the first user 211.

[0227] This completes the execution of the second method 900.

[0228] Thus, some non-limiting embodiments of the second method 900 are intended to provide highly individualized recommendations regarding whether to allow or disallow receiving third party cookies to various users of the communications network 240.

[0229] Possible changes and improvements to the above-described embodiments of the present technology may be apparent to one skilled in the art. The foregoing description is provided for illustrative purposes only and is not intended to limit the scope of the invention. The scope of protection of this technology is determined solely by the scope of the attached claims.

[0230] Although the embodiments described above are given with reference to specific steps performed in a particular order, it should be understood that these steps may be combined, separated, or their order may be changed without departing from the scope of the present technology. Accordingly, the order and grouping of steps are not restrictive for this technology.

Claims (101)

1. A computer-based method for managing the tracking of a target user's web browsing operations in a browser application, implemented by an electronic device executing the browser application and connected to a server executing a machine learning algorithm (MLA) trained by the server to determine probability values for the usage stage user's permission to use cookies. third party files from the third party web resources of the use phase together with the third party web resources when viewing the web pages of the use phase containing elements associated with the third party web resources of the use phase, wherein the MLA algorithm is trained on the training data set, containing many learning objects, each of which contains: (a) third party learning web resource-dependent data of the third party learning web resource associated with the learning web page; (b) training user-specific data of the training user viewing the training web page, including data about the training user's past user activity on a third party training web resource; And (c) a label indicating whether or not a training user is allowed to use a third party training cookie from a third party training web resource in conjunction with a third party training web resource when viewing a training web page containing elements related to the training website. third party resource wherein the third party cookie was previously received by the training user's electronic learning device; the method includes: - receipt by an electronic device from this web server of data representing the web page of the use stage for display in a browser application; - determination by an electronic device, based on data representing the use stage web page, of its elements associated with at least one third party web resource of the use stage, wherein such element is capable of sending data about actions with it to the third party web server containing at least a third party web resource other than this web server; - receipt by an electronic device of use-phase data containing (a) user-dependent data of the target user, including data about the target user's past user activities with at least one third party web resource of the use-phase, and/or (b) dependent on the web third party resource data of at least one third party web resource of the use stage; - input by the electronic device of the use-phase data into the MLA algorithm to determine the probability value of the target user allowing the use of a third-party use-phase cookie from at least one third-party web resource of the use-phase in conjunction with at least one third-party web resource of the step usage while browsing a usage stage web page, wherein a third party usage stage cookie has previously been received by the target user's electronic device and enables the third party server to track the target user's web browsing activities in the browser application and the sharing permission probability values a third party usage stage cookie from at least one third party usage stage web resource is different for the target user and for another target user; - determining by the electronic device that it is unlikely that the target user will consent to the use of a third party cookie of the use phase in conjunction with at least one third party web resource of the use step while viewing a web page of the use step, if the probability value is less than a probability threshold; And - generation by the electronic device of a first notification advising the target user not to use the third party cookie of the use phase in conjunction with at least one third party web resource of the use phase. 2. The method of claim 1, wherein if the probability value is less than a probability threshold value, the method further includes causing the electronic device to prevent the browser application from sharing the third party cookie with at least one third party web resource of step use. 3. The method according to claim 1, characterized in that if the probability value is greater than or equal to the threshold probability value, the method additionally includes: - determining by the electronic device the likely permission of the target user to use the third party cookie in conjunction with at least one third party web resource of the use phase; And - generation by the electronic device of a second notification notifying about the safety of using a third-party cookie in conjunction with at least one third-party web resource at the use stage. 4. The method according to claim 3, characterized in that it additionally includes: - receipt by the electronic device of an indication of a user action of the target user performed in response to the first notification or second notification; And - storing by the electronic device an indication of the user action for the purpose of updating the training data set for training the MLA algorithm if the user action does not correspond to the first notification or the second notification. 5. The method of claim 1, wherein if the probability value is greater than or equal to a probability threshold value, the method further includes causing the electronic device to share a third party cookie from at least one third party web resource with the browser application. sides of the use stage. 6. The method according to claim 1, characterized in that the data of the third party educational web resource contains the domain zone of the third party educational web resource, and/or the domain name of the third party educational web resource, and/or the category of the third party educational web resource sides. 7. The method according to claim 6, characterized in that the following categories of third-party educational web resource are used: (a) social network, and/or (b) news portal, and/or (c) video hosting platform. 8. The method according to claim 1, wherein the user-specific data of the target user further comprises data about the current location of the target user. 9. The method according to claim 1, characterized in that the server is connected to another electronic device of another user, other than the electronic device, which runs another browser application and is capable of: - receive data from this web server representing the web page of the use stage for display in another browser application; - determine, based on data representing the use stage web page, its elements associated with at least one third party web resource of the use stage; - receive other usage stage data containing (a) other user-dependent data of another user, including data about another target user's past user activities with at least one third party web resource of the usage stage, and/or (b) web-dependent - third party resource data of at least one third party web resource of the use stage; - input other usage phase data into the MLA algorithm to determine a different probability value of allowing another target user to use the usage phase third party cookie in conjunction with at least one usage phase third party web property while viewing a usage phase web page, when This is a third-party usage-stage cookie that has previously been received by another target user's other electronic device and provides the third party server with the ability to track the other user's web browsing activity in another browser application and a different probability of allowing the other target user to use the cookie. the third party of the use phase in conjunction with at least one third party web resource of the use phase is different from the corresponding probability value determined for the target user; - determine the likelihood of another target user authorizing the use of a third party cookie of the use step in conjunction with at least one third party web resource of the use step while viewing a web page of the use step, if the other probability value is less than the probability threshold; And - generate a first notice advising the other target user to opt out of using a third-party usage-stage cookie in conjunction with at least one third-party usage-stage web property. 10. The method according to claim 1, characterized in that the MLA algorithm is an MLA algorithm based on CatBoost decision trees. 11. A computer-based method for managing tracking of a target user's web browsing operations in a browser application, implemented by an electronic device executing the browser application and connected to a server executing an MLA algorithm trained by the server based on a training dataset containing a plurality of training objects to determine resolution probability values user of the use stage to receive third party cookies from third party web resources of the use stage when viewing related web pages of the use stage, wherein the learning object of the plurality of learning objects contains: (a) third party learning web resource-dependent data of the third party learning web resource associated with the learning web page; (b) training user-specific data of the training user viewing the training web page, including data about the training user's past user activity on a third party training web resource; And (c) a label indicating whether or not the training user is allowed to receive a third party training cookie from the third party training web resource when viewing the training web page; the method includes: - receipt by an electronic device from this web server of data representing the web page of the use stage for display in a browser application; - determination by an electronic device, based on data representing the use stage web page, of its elements associated with at least one third party web resource of the use stage, wherein such element is capable of sending data about actions with it to the third party web server containing at least a third party web resource other than this web server; - receipt by an electronic device of use-phase data containing (a) user-dependent data of the target user, including data about the target user's past user activities with at least one third party web resource of the use-phase, and/or (b) dependent on the web third party resource data of at least one third party web resource of the use stage; - input by the electronic device of the use-phase data into the MLA algorithm to determine a probability value of the target user's permission to receive from at least one third-party web resource the use-phase of a third-party use-phase cookie while viewing a use-phase web page, wherein the cookie - The third-party use-phase file enables the third-party server to track a target user's web browsing activity in a browser application and the probability values of allowing a third-party use-phase cookie from at least one third-party use-phase web resource for the target user. and for another target user are different; - determining by the electronic device that it is unlikely that the target user will authorize the use-phase third-party cookie from at least one use-phase third-party web resource while viewing a use-phase web page if the probability value is less than a probability threshold; And - generation by the electronic device of a first notification advising the target user not to receive a third-party use-phase cookie from at least one third-party use-phase web resource. 12. The method of claim 11, wherein if the probability value is less than a probability threshold value, the method further includes causing the electronic device to refuse the browser application to receive a third party cookie from the at least one third party web resource of step use. 13. The method according to claim 11, characterized in that if the probability value is greater than or equal to the threshold probability value, the method further includes: - determining by the electronic device the likely permission of the target user to receive a third party cookie from at least one third party web resource of the use phase; And - generation by the electronic device of a second notification notifying about the security of permission to receive a third party cookie from at least one third party web resource of the use stage. 14. The method according to claim 13, characterized in that it additionally includes: - receipt by the electronic device of an indication of a user action of the target user performed in response to the first notification or second notification; And - storing by the electronic device an indication of the user action for the purpose of updating the training data set for training the MLA algorithm if the user action does not correspond to the first notification or the second notification. 15. The method of claim 11, wherein if the probability value is greater than or equal to a probability threshold value, the method further includes causing the electronic device to allow the browser application to receive a third party cookie from the at least one web resource third party use stage. 16. The method according to claim 11, characterized in that the third party educational web resource data contains the domain zone of the third party educational web resource and/or the domain name of the third party educational web resource and/or the category of the third party educational web resource. 17. The method according to claim 16, characterized in that the following categories of third-party educational web resource are used: (a) social network, and/or (b) news portal, and/or (c) video hosting platform. 18. The method of claim 11, wherein the user-specific data of the target user further comprises data about the current location of the target user. 19. The method according to claim 11, characterized in that the server is connected to another user's electronic device other than the electronic device, which runs another browser application and is capable of: - receive data from this web server representing the web page of the use stage for display in another browser application; - determine, based on data representing the use stage web page, its elements associated with at least one third party web resource of the use stage; - receive other usage stage data containing (a) other user-dependent data of another user, including data about another target user's past user activities with at least one third party web resource of the usage stage, and/or (b) web-dependent - third party resource data of at least one third party web resource of the use stage; - input other usage phase data into the MLA algorithm to determine a different probability value of allowing another target user to receive a third party usage phase cookie from at least one third party web resource while viewing a usage phase web page, wherein a third-party usage-stage cookie enables a third-party server to track web browsing activity performed by another user in another browser application, and another value for the probability of allowing a third-party usage-stage cookie from at least one third-party web resource the use stage for another target user is different from the corresponding probability value determined for the target user; - determine the likelihood of another target user authorizing a use-phase third party cookie from at least one use-phase third-party web resource while browsing a use-phase web page if the other probability value is less than the likelihood threshold; And - generate a first notification advising the other target user to opt out of receiving a third-party usage-stage cookie from at least one third-party usage-stage web resource. 20. The method according to claim 11, characterized in that the MLA algorithm is an MLA algorithm based on CatBoost decision trees. 21. A computer method for managing tracking of a user's web browsing operations in a browser application, implemented by the processor of an electronic device executing the browser application, and including: - receipt by the processor from this web server of data representing a web page for display in a browser application; - determination by the processor, based on the data representing the web page, of its elements associated with at least one third party web resource, wherein such element is capable of sending data about actions with it to the third party web server containing at least one web resource a third party resource other than this web server; - receipt by the processor of (a) user-dependent user data, including data about the user's past user actions with at least one third party web resource; and (b) third party web property-dependent data of at least one third party web property; - input by the processor of user-specific data and third-party web resource-dependent data into the MLA algorithm to determine the probability value of the user allowing the use of a third-party cookie in conjunction with at least one third-party web resource while browsing a web page, wherein the third party cookie of the usage stage has previously been received by the user's electronic device, and the third party cookie provides the third party server with the ability to track the user's web browsing operations in the browser application, while the MLA algorithm is trained to determine the probability values of the user's permission to use the cookie - third party cookies in conjunction with third party web properties when viewing associated web pages based on at least user-specific data, and the probability value of allowing the use of a third party cookie in conjunction with at least one third party web resource are different for a user and for another user; - determining by the processor that it is unlikely that the user will authorize the use of a third party cookie in conjunction with at least one third party web resource of the use phase while browsing a web page, if the probability value is less than a probability threshold; And - generation by the processor of a first notification recommending that the user refuse to use a third-party cookie in conjunction with at least one third-party web resource. 22. The method of claim 21, wherein if the probability value is less than a probability threshold, the method further includes causing the processor to prevent the browser application from sharing the third party cookie with the at least one third party web resource. 23. The method according to claim 21, characterized in that if the probability value is greater than or equal to the threshold probability value, the method further includes: - determination by the processor of the user's likely permission to use the third party cookie in conjunction with at least one third party web resource; And - generation by the processor of a second notification notifying about the safety of using a third-party cookie in conjunction with at least one third-party web resource. 24. The method according to claim 23, characterized in that it additionally includes: - receipt by the processor of an indication of a user user action performed in response to the first notification or second notification; And - storing by the processor an indication of a user action associated with the at least one third party web resource for the purpose of updating the training data set used to train the MLA algorithm if the user action does not correspond to the first notification or the second notification. 25. The method of claim 21, wherein if the probability value is greater than or equal to a probability threshold, the method further includes causing the processor to use the third party cookie by the browser application in conjunction with at least one third party web resource . 26. The method of claim 21, wherein the user-specific data comprises (a) data about the user's current location and/or (b) data about the user's past user activities with at least one third party web resource. 27. The method according to claim 26, characterized in that the data about the user's past user activities with at least one third party web resource comprises (a) the number of visits by the user to at least one third party web resource during a predetermined period and /or (b) the total number of visits by the user to at least one third party web resource. 28. The method according to claim 21, characterized in that the data dependent on the third party web resource contains (a) a domain zone of at least one third party web resource, and/or (b) a domain name of at least one web resource third party resource, and/or (c) a category of at least one third party web resource. 29. The method according to claim 28, characterized in that the following categories of third-party educational web resource are used: (a) social network, and/or (b) news portal, and/or (c) video hosting platform. 30. The method according to claim 21, characterized in that the MLA algorithm is trained by a server associated with an electronic device based on a training data set containing a plurality of training objects, each of which contains: (a) third party learning web resource-dependent data of the third party learning web resource associated with the learning web page; (b) training user-dependent data of the training user viewing the training web page, including consent or opt-out of receiving data from the training user's past user activities with a third party training web resource; And (c) a label indicating whether or not a training user is allowed to use a third party training cookie from a third party training web resource in conjunction with a third party training web resource when viewing a training web page containing elements related to the training website. a third party resource, wherein the third party cookie was previously received by the learning electronic device of the learning user. 31. The method according to claim 30, characterized in that the data of the third party educational web resource contains the domain zone of the third party educational web resource, and/or the domain name of the third party educational web resource, and/or the category of the third party educational web resource sides. 32. The method according to claim 21, characterized in that the user and the other user view the web page of the use stage from different electronic devices. 33. The method according to claim 21, characterized in that the MLA algorithm is an MLA algorithm based on CatBoost decision trees. 34. An electronic device for managing the tracking of a target user's web browsing activities in a browser application of the electronic device associated with a server executing an MLA algorithm trained by the server to determine probability values for the usage stage user's permission to use third party cookies from third party web properties. use stage in conjunction with third party web resources when viewing use stage web pages containing elements associated with third party use stage web resources, wherein the MLA algorithm is trained on a training dataset containing a plurality of training objects, each of which contains : (a) third party learning web resource-dependent data of the third party learning web resource associated with the learning web page; (b) training user-specific data of the training user viewing the training web page, including data about the training user's past user activity on a third party training web resource; And (c) a label indicating whether or not a training user is allowed to use a third party training cookie from a third party training web resource in conjunction with a third party training web resource when viewing a training web page containing elements related to the training website. a third party resource, wherein the third party cookie was previously received by the educational electronic device of the educational user, and the electronic device includes a processor and a machine-readable physical storage medium storing instructions, and the processor, when executing the instructions, is capable of: - receive data from this web server representing the web page of the usage stage for display in the browser application; - determine, based on data representing the use stage web page, its elements associated with at least one third party web resource of the use stage, wherein such element is capable of sending data about actions with it to a third party web server containing at least to the extent a third party web resource other than this web server; - obtain usage stage data containing (a) user-specific data of the target user, including data about the target user's past user activities with at least one third party web resource of the usage stage, and/or (b) dependent on the third party web resource party data of at least one third party web resource of the use stage; - input the usage phase data into the MLA algorithm to determine a probability value of the target user allowing the use of a third party usage phase cookie from at least one third party web resource of the usage phase in conjunction with at least one third party web resource of the usage phase in browsing time of a usage stage web page, wherein the third party cookie of the usage stage is previously received by the target user's electronic device and provides the third party server with the ability to track the target user's web browsing operations in the browser application, and the cookie sharing permission probability values - a use stage third party file from at least one use stage third party web resource for a target user and for another target user are different; - determine the likelihood of the target user consenting to the use of a third party cookie of the use stage in conjunction with at least one third party web resource of the use stage while viewing a web page of the use stage, if the probability value is less than the probability threshold; And - generate a first notice advising the target user not to use a third-party usage-stage cookie in conjunction with at least one third-party usage-stage web property. 35. The electronic device of claim 34, wherein the processor is further capable of causing the browser application to prevent the use of the third party cookie in conjunction with the at least one third party web resource of the use step if the probability value is less than a probability threshold. 36. The electronic device according to claim 34, characterized in that if the probability value is greater than or equal to the threshold probability value, the processor is additionally capable of: - determine the likely permission of the target user to use the third party cookie in conjunction with at least one third party web resource of the use phase; And - generate a second notification indicating the safety of using a third party cookie in conjunction with at least one third party web resource of the use stage.