TWI771602B - Guiding system based on prediction behavior and method thereof - Google Patents

Abstract

A guiding system based on predictive behavior is provided. When a user operates a terminal device to visit an official bank website or execute a banking application, the unique individual identifier (UIID) of the user is identified. A session ID corresponding to the UIID of the user is obtained. Behavior data of the user is continuously recorded according to the session ID and the UIID of the user. The behavior data is classified according to preset conditions, and then the classification result is stored as historical classification data corresponding to the UIID of the user. The historical classification data corresponding to the UIID of the user is founded and output. Predicted transaction data of the user is generated according to the historical classification data and/ or the behavior data corresponding to the UIID of the user, so as to provide a push notification or a pop-up window corresponding to the predicted transaction data to the terminal device. Therefore, the guiding system based on predictive behavior can achieve the technical effect of improving the experience of the user.

Description

Predictive behavior-based guidance system and method therefor

The present invention relates to a guidance system and method thereof, in particular to a guidance system and method based on predicted behavior.

With the advancement of science and technology, the popularization of education, and the development of social economy, people have a very high willingness to accept and operate technology products. topics of concern.

At present, each bank provides its official website or application (Application, App) to provide users with information pages of relevant business items, and users usually need to click or search for the information they need. It can be seen that the current information provision method relies more on the user's own operation, which is obviously not efficient and not intelligent enough.

To sum up, it can be seen that there has been a long-standing problem in the prior art that the way in which the bank provides information relies on the user's self-operation. Therefore, it is necessary to propose an improved technical means to solve this problem.

The invention discloses a guidance system and method based on predicted behavior.

First of all, the present invention discloses a guidance system based on predictive behavior, which is used to provide a personalized push notification or pop-up window to the terminal device when a user operates a terminal device to visit an official bank webpage or execute a bank application program. The guidance system based on predicted behavior includes: identity recognition module, access module, behavior record module, classification module, data module and business logic module. The identity identification module is used to identify the user's login identification profile (Unique Individual Identifier, UIID) when the terminal device starts to visit the official website of the bank or execute the bank application; the access module is used to obtain the number plate (Session ID) to Corresponding to the user's login identification profile, wherein the number plate is random and not reused; the behavior record module is used to continuously receive and store the terminal device according to the number plate and login identification profile from the beginning of visiting the bank's official website to leaving the bank's official website. , from the execution of the bank application to the closing of the bank application, or the behavior data from the start of visiting the bank's official website or the execution of the bank application to staying on a single page for more than a preset time; the classification module connected to the behavior record module is used to continuously receive Behavior data, and classify the behavior data according to the preset conditions to output the behavior classification data; the data module connecting the classification module and the identity recognition module is used to continuously receive the behavior classification data from the classification module and store it as a corresponding Log in the historical classification data of the identification profile, and find and output the corresponding historical classification data according to the login identification profile; the business logic module connected to the data module is used to generate the predicted transaction data of the user according to the historical classification data , to provide a pop-up window or push notification corresponding to the predicted transaction data to the terminal device.

In addition, the present invention discloses a guidance method based on predicted behavior, which is used for providing a personalized push notification or pop-up window to the terminal device when a user operates a terminal device to visit an official bank webpage or execute a bank application program. The steps include: when When the terminal device starts to visit the official website of the bank or execute the bank application, it recognizes the user's login identification profile; obtains a number plate to correspond to the user's login identification profile, wherein the number plate is random and not reused; according to the number plate And the login identification profile continues to receive and store the terminal device from the start of visiting the bank's official website to leaving the bank's official website, from executing the bank's application to closing the bank's application, or from the beginning of visiting the bank's official website or executing the bank's application to staying on a single page. Behavior data of a preset time; continue to receive behavior data, and classify the behavior data according to preset conditions to output behavior classification data; continue to receive behavior classification data to store as historical classification data corresponding to the login identification profile, and based on Log in the identification profile to find and output the corresponding historical classification data; and generate the user's predicted transaction data according to the historical classification data and/or the behavior data, so as to provide a pop-up window or push broadcast corresponding to the predicted transaction data Notify the terminal device.

The system and method disclosed in the present invention are as above, and the difference from the prior art is that the present invention identifies the user's login identification profile when the user operates the terminal device to access the bank's official website or executes the bank application program, and obtains the corresponding login identification profile. continuous recording of the user's behavior data according to the number plate and the login identification profile; classify the behavior data according to the preset conditions and store it as historical classification data corresponding to the login identification profile; find and output the corresponding login identification profile according to the login identification profile The historical classification data of the user; the predicted transaction data of the user is generated according to the historical classification data and/or the behavior data, so as to provide a pop-up window or push notification corresponding to the predicted transaction data to the terminal device.

Through the above-mentioned technical means, the present invention can achieve the technical effect of improving user experience.

Before describing the predictive behavior-based guidance system and the method disclosed in the present invention, the terms defined by the present invention will be described first, the modules included in the predictive behavior-based guidance system of the present invention and the back-end of the bank The system can be implemented in various ways, including software, hardware, firmware, or any combination thereof. For example, in some embodiments, the mobile device may be implemented using software and/or hardware, and the scope of the present invention is not limited in this regard. Software or firmware using the techniques proposed in the implementation may be stored on machine-readable storage media, such as: read only memory (ROM), random access memory (RAM), disk storage media, optical storage media, flash storage media Flash memory devices, etc., and may be executed by one or more general purpose or special purpose programmable microprocessors.

The modules described in the present invention can be connected to each other in a wired or wireless manner, so as to transmit information and data. The predictive behavior-based guidance system of the present invention and the bank's back-end system can communicate with each other through a network communication system, such as a mobile communication network, the Internet, a local area network, a wide area network and/or a wireless network, so as to To transmit information and data.

The embodiments of the present invention will be described in detail below in conjunction with the drawings and examples, so as to fully understand and implement the implementation process of how the present invention applies technical means to solve technical problems and achieve technical effects.

Please refer to "Fig. 1" and "Fig. 2" first, "Fig. 1" is a system block diagram of an embodiment of the guidance system based on predictive behavior of the present invention, and "Fig. 2" is the basis of "Fig. 1" A method flow chart of an embodiment of a predictive behavior-based guidance system for executing a predictive behavior-based guidance method. The predictive behavior-based guidance system 100 can be used to provide a personalized push notification or pop-up window to the terminal device to improve user experience when a user operates a terminal device (not shown) to visit an official bank webpage or execute a bank application.

In this embodiment, the guidance system 100 based on predicted behavior may include: an identity recognition module 110 , an access module 120 , a behavior record module 130 , a classification module 140 , a data module 150 and a business logic module 160 . The classification module 140 can be connected to the behavior recording module 130 , the data module 150 can be connected to the classification module 140 and the identity recognition module 110 respectively, and the business logic module 160 can be connected to the data module 150 .

The identification module 110 can be used to identify the user's login identification profile when the terminal device starts to visit the official website of the bank or execute the bank application (ie, step 210 ). More specifically, when the user operates the terminal device to visit the official bank website or execute the bank application in the logged-in state, the user's account number is used as the user's login identification profile (ie the user's identity); and when the user operates When the terminal device is not logged in to visit the official website of the bank or execute the bank application, the user's login identification profile is identified according to the browsing information (Cookie) stored in the terminal device. In other words, the identity recognition module 110 can use the user's account or the browsing information stored in the terminal device to recognize the user's login profile.

The access module 120 can be used to obtain a session ID corresponding to the user's login identification profile, wherein the session ID is random and not reused (ie, step 220 ). In more detail, the number plate is a serial number that does not repeat. When the identity recognition module 110 recognizes the user's login identification profile and the access module 120 obtains the number plate corresponding to the user's login identification profile, based on the predicted behavior The guidance system 100 can know the login identification profile of the user who visits the official bank website or executes the bank application program and the number plate obtained. Therefore, the guidance system 100 based on predicted behavior can learn from the login identification profile of the user and the number plate obtained by the user. The number of number plates obtained, and the number of times users visit the bank's official website or execute the bank's application. If the predictive behavior-based guidance system 100 records the time each time the user visits the official bank website or executes the bank application, the frequency of the user visiting the bank official website or executing the bank application can be known.

The behavior record module 130 can be used to continuously receive and store the terminal device according to the number plate and the login identification profile from the beginning of visiting the bank's official website to leaving the bank's official website, from executing the bank application to closing the bank's application, or from the beginning of visiting the bank's official website. Or execute the bank application to stay on a single page for more than a preset time of behavior data (ie, step 230 ). In other words, the behavior record module 130 can be used to record the browsing history of the user visiting the bank's official website or executing the bank application (when staying on a single page for more than a preset time, the behavior record module 130 can automatically stop recording the browsing history) . The browsing process includes a plurality of behavior data, that is, the behavior recording module 130 can record the browsing process in segments according to the set time length (each segment of the browsing process corresponds to one behavior data), and each behavior data can include at least one operation behavior, The latitude and longitude where the terminal device is located, the execution time of each operation behavior, the type of the terminal device, the operating system version of the terminal device, and the browser version of the terminal device. For example, the operation behavior can be click on the exchange rate spreadsheet or search for words on the site , the preset time and the set time length can be adjusted according to actual needs.

The classification module 140 can be used to continuously receive the behavior data, and classify the behavior data according to a preset condition to output the behavior classification data (ie, step 240 ). In this embodiment, since the behavior record module 130 can be used to record the browsing history of the user visiting the official bank website or executing the bank application, the user's identity can be identified in two ways: the user's account number or the user's account stored in The browsing information of the terminal device, therefore, the guidance system 100 based on the predicted behavior can classify the corresponding behavior data according to different preset conditions based on the user's account or the browsing information stored in the terminal device, respectively, to output the behavior classification data. .

When the login identification profile that identifies the user is based on the user's account number, the classification module 140 can classify the behavioral data according to the characteristics of the thing. For example, the preset conditions may include five items: "form", "visiting individual", "individual region", "transaction target" and "search target", wherein, "form" is for users to operate the terminal device to apply for business The classification items of the filling process can include details such as "Form Recognition", "Field Recognition", "Field Interaction" and "Form Sending"; It can include details such as "user's login identification profile" and "individual attribute"; "individual region" is the classification item of the user's location and time when operating the terminal device, and can include "the IP address of the terminal device", Details such as "latitude and longitude where the terminal device is located", "city where the terminal device is located", and "execution time of each operation behavior"; "transaction target" is a classification item of bank marketing or tradable target, which may include "foreign currency" , "Foreign Exchange" and "Savings Insurance"; "Search Target" is the classification item of the user's search target, which may include "Search Media (ie operating system version, browser version and/or vehicle type)" However, this embodiment is not intended to limit the present invention, and the preset conditions can be adjusted according to actual needs. In this embodiment, the classification module 140 can classify each behavior data of the user's account based on the five major items and its detailed items, so as to output a corresponding behavior classification data. Generally speaking, each behavioral data must include data corresponding to "visiting individuals" and "individual regions", and whether to include data corresponding to "forms", "transaction targets" and "search targets" depends on operational behaviors and predictions. Depends on the setting of the conditions.

When the login identification profile for identifying the user is based on the browsing information stored in the terminal device, the classification module 140 can first set the context attribute of each behavior data according to the "system event" and "customized event", and then according to the " Behavior Objects, Internal Search Objects, Target Objects, Push Objects, and Outline Objects are classified to correspond to output behavior classification data. Each setting value can include multiple items. , for example: "Behavior Object" can include details such as "Page Load", "Variable ID" and "Variable Value", and "Internal Search Object" can include details such as "Page Load" and "Internal Search Name Attribute" , "Target Object" may include details such as "Page Load", "Field Interaction", "Field Value", "Object ID" and "Target Name", and "Outline Object" may include "User's Login ID" Outline and other details, and "Push Object" can include such details as "Object Hierarchy Number", "Object Link", "Object Type", "Push Name", "Push Location" and "Push Category". In other words, in this embodiment, the preset conditions include two major items (namely, “system events” and “customized events”) and five setting values and their detailed items, but this embodiment is not intended to limit the The invention can adjust the preset conditions according to actual needs. The classification result of this preset condition can be used by website administrators or product designers to further analyze the user's intention. , you need to find the page loading and internal search name attributes in the system event (that is, the object number or name of the page search), the information flow will continue to stream the matching conditions to the search object, and then you can perform keyword and search number of times presented.

It should be noted that, in this embodiment, based on the user's account or the browsing information stored in the terminal device, different preset conditions are used to classify the corresponding behavior data, which can be used for different customers (existing customers, namely The user's account is used as the user's login identification profile, and the new account (that is, the browsing information stored in the terminal device is used as the user's login identification profile) to classify the behavior data for subsequent data analysis, and then target the official webpage or application. The design is adjusted to enhance the user (existing and/or new) experience.

The data module 150 can continuously receive the behavior classification data, store it as historical classification data corresponding to the login identification profile, and find and output the corresponding historical classification data according to the login identification profile (ie, step 250 ). In other words, the data module 150 can output the historical classification data corresponding to the user's login identification profile to the business logic module 160 .

The business logic module 160 can be used to generate the predicted transaction data of the user according to the historical classification data and/or the behavior data, so as to provide a pop-up window or push notification corresponding to the predicted transaction data to the terminal device (ie, step 260 ). . In other words, the business logic module 160 can use the user's past and current browsing history to estimate the user's next possible operation behavior, generate the user's predicted transaction data, and guide the user through push notifications or pop-up windows. The user is directed to the relevant page of the expected operation behavior estimated by the business logic module 160, so as to improve the user's experience and solve the inefficiency caused by the fact that the method of providing information by the bank relies on the user's own operation in the prior art. Too high is not smart enough. The method for the business logic module 160 to use the user's past and current browsing history to estimate the user's expected operation behavior may include the following three methods.

The first method is as follows: the preset condition may include a transaction target item, and the business logic module 160 may count the number of occurrences of different transaction targets in the historical classification data corresponding to the user's identity according to the transaction target item, so as to adjust each transaction target item. The weighted index corresponding to a trading target (that is, the more times the trading target appears, the greater the weighted index corresponding to the trading target). The window method guides the user to the corresponding page of the transaction target (that is, the content of the pop-up window or push notification includes the link of the page corresponding to the transaction target, so that the user can click the link to navigate to the page). It should be noted that, since the preset conditions may include transaction target items, this method is suitable for page guidance when the user operates the terminal device to access the official website of the bank or execute the bank application in the logged-in state.

In addition, this method can also be applied to the page guidance when the user operates the terminal device next time to visit the official website of the bank or execute the bank application program. More specifically, the business logic module 160 can find and store the highest occurrence/ The page of the transaction target with the largest weighted index enables the predictive behavior-based guidance system 100 to directly lead to this page when the user operates the terminal device next time to visit the official bank website or execute the bank application (that is, the predictive behavior-based guidance After the user operates the terminal device next time to visit the bank's official website or execute the bank application program, the system 100 first performs the user's identification, and then directly guides the screen of the terminal device to the highest number of occurrences/maximum weighting index. the transaction target page).

The second method is as follows: the guidance system 100 based on the predicted behavior may further include a learning module 170, which connects the behavior recording module 130 and the business logic module 160, so as to correspond to the user's login identification profile. The behavior data obtains the top-N browsing paths more frequently used by the user, so that the business logic module 160 generates the user's predicted transaction data (ie, business logic) according to the top-N browsing paths more frequently used by the user. The module 160 can be used to generate the predicted transaction data of the user according to the behavior data, so as to provide a pop-up window or push notification corresponding to the predicted transaction data to the terminal device, wherein N is a positive integer greater than zero. In other words, the business logic module 160 can be loaded according to the top-N browsing path and the current screen of the terminal device (which can be based on the transaction target or page load included in the latest historical classification data or behavior data) that is commonly used by the user. and other detailed judgments) to evaluate the user's possible next operation behavior, and then find the corresponding page according to the evaluation results, so as to include the link of the page in the content of the pop-up window or push notification for the user to click on the link and lead to this page. It should be noted that this method is suitable for page guidance when the user operates the terminal device to access the official website of the bank or execute the bank application in the logged-in or logged-in state.

The third method can be combined with the first method and the second method, that is, the business logic module 160 can be based on the number of occurrences of different transaction objects in the historical classification data corresponding to the user's login identification profile and the more commonly used The top N browsing paths generate the predicted transaction data of the user, so as to provide a pop-up window or push notification corresponding to the predicted transaction data to the terminal device. It should be noted that this method is suitable for page guidance when the user operates the terminal device to access the official website of the bank or execute the bank application in the logged-in state.

Through the above steps, users can predict the target object they want to use according to the user's current or past operation behavior of visiting the bank's official website or executing the bank's application, and provide relevant pages to the user, so that the user can perform the least operation steps. (ie, within the shortest path time), direct the display page of the terminal device operated by the user to the page of the target object, so as to avoid the possibility of interruption and abandonment of use by the user due to complicated, lengthy and cumbersome operation procedures, and Improve usage rate and good user experience.

In this embodiment, when the predicted transaction data includes business items that need to be handled over the counter, the business logic module 160 can be connected to the bank's back-end system (not shown) and based on the terminal device included in the latest historical classification data (because the behavior record module 130 continuously records the browsing history, and each behavior data must include the data corresponding to the "individual region", therefore, the latitude and longitude of the terminal device of the newly generated historical classification data (that is, the terminal The location information of the device) is the location of the current terminal device) to find the service bases near the terminal device, the current number of people waiting for each service base, the certificates and the required time for handling the business, so that the push broadcast The notification or the displayed content of the pop-up window includes the business item that needs to be handled at the counter, the service locations near the terminal device, the current number of people waiting at each service location, the certificates required for the business, and the required time. Among them, the bank's back-end system is a system that grasps the address of each service point and the number of people waiting in the current service hall, and has a database of the average time spent for each transaction of the bank and the required documents. Therefore, the business logic module 160 can connect to the bank's back-end system to obtain relevant information for evaluating the business that the user wants to perform over-the-counter (the certificates and time required to handle the business) and find services near the terminal device. Base, the current number of people waiting at each service base, so that the content displayed in the push notification or the pop-up window includes the business item that needs to be handled at the counter, the service bases near the terminal device, the current number of people waiting at each service base, and When the business logic module 160 evaluates that the business that the user wants to perform over-the-counter is correct, it can allow the user to decide whether to go to a nearby service location to handle the business at the moment, and then provide Good user experience.

In addition, the display content of the push notification or pop-up window can also provide the user with the option of whether to go to a nearby service location to handle the business, and if the user chooses to go to a nearby service location to handle the business, the user can get an online number service; If users give up going to a nearby service location to handle the business, follow-up data analysis can be provided to determine whether many users give up on-the-counter processing, so that the bank may consider the possibility of changing the processing of the business to online processing.

In this embodiment, in order to improve the user's experience, the guidance system 100 based on the predicted behavior can use the above-mentioned predicted user's operation behavior, and can also use the historical classification data of existing customers and/or new customers and/or The data analysis of browsing history (that is, behavioral data), to obtain the design of the official website or application can be adjusted to further enhance the user experience. In this embodiment, based on the historical classification data and/or browsing history (ie behavior data) of existing customers and/or new customers, the conversion of traffic sources, vehicles, traffic, function pages, latest news, and advertisement clicks and transactions can be performed. The detailed analysis is as follows, but the actual data analysis items can be adjusted according to actual needs, and the traffic sources, vehicles, traffic, function pages, latest news and advertisement clicks in this embodiment are not related to Analysis of transaction conversion rates is limited.

In the part of traffic source analysis, each behavior data can include the latitude and longitude of the terminal device and the data of the execution time of each operation behavior. According to the data corresponding to the execution time of each operation behavior in the behavior data, it can be determined whether the user is Staying on some pages for too long means that the user is highly interested in a certain product or the setting process of the page is not good. For example, the product application page fills in too much information, resulting in a poor user experience. It can be provided to Developers and product managers of official webpages or applications are used as data for future optimization of products and official webpages or applications.

In the part of vehicle analysis, because the types, operating system versions and browser versions of current terminal devices are too diverse, for example: the operating system version can be Windows or Mac; the browser version can be IE, Chrome, Firefox or Safari; The type can be a computer, smart phone or tablet, which makes it difficult for developers of official web pages or applications to quickly obtain information about the terminal device operated by the user when troubleshooting problems. Therefore, when each behavioral data record contains When the type, operating system version and browser version of the terminal device operated by the user, by analyzing the data and analyzing the related questions obtained by other data analysis, the information of the terminal device operated by the user and its information can be understood. Problems encountered so that developers of official web pages or applications can troubleshoot related problems.

In the part of traffic analysis, each behavior data can include the latitude and longitude of the terminal device and the data of the execution time of each operation behavior. According to the data corresponding to the "execution time of each operation behavior" in the behavior data, the data can be analyzed by date The overall traffic and average traffic of each period, and identify the peak date and peak period of the official website or application, which can be used as the basis for resource diversion from the perspective of historical statistics. Load traffic of web pages or applications.

In the part of functional page analysis, the top M browsing paths most frequently used by all users can be calculated based on the browsing history of all users (ie existing customers and new users), which can be used by developers of official web pages or applications in the future. Adjust the basis for page design. Among them, M is a positive integer greater than zero.

In the section of latest news and advertisement click analysis, you can obtain the latest news and pages of product promotions on the shelves of each business based on the historical classification data and/or browsing history (ie behavior data) of all users (ie, existing customers and new customers). The number of attentions in the advertisement area, and the setting of the target object in the classification module 140 is used to collect the number of clicks of each target object within a certain time interval, and then obtain the operation page of the high frequency clicks of each target object. Therefore, it can provide a reference for product managers to design new products in the future.

In the part of transaction conversion rate analysis, data analysis can be performed based on the historical classification data and/or browsing history (ie behavior data) of existing customers to obtain the data corresponding to the "form" items. service, the payment process of the entire application service includes filling in the application page, channel authorization verification, authorization reply, and returning the application result. If any of these steps are suspended, the transaction may fail. Therefore, it can be classified based on the history of each existing customer. Data and/or browsing history (i.e. behavioral data) to obtain the number of suspensions (i.e. not completed) and completions for each stage, and use the number of suspensions and completions for each stage to obtain the conversion rate for each stage (each stage The conversion rate for each stage is the ratio of aborts to completions) to get an idea of the conversion rate for each stage.

To sum up, it can be seen that the guidance system based on the predicted behavior of the present invention recognizes the user's login identification profile when the user operates the terminal device to visit the bank's official website or executes the bank application program; obtains a number plate corresponding to the login identification profile; The number plate and the login identification profile continuously record the behavior data of the user; classify the behavior data according to the preset conditions, and store it as historical classification data corresponding to the login identification profile; find and output the corresponding historical classification according to the login identification profile data; according to the historical classification data and/or the behavior data, the predicted transaction data of the user is generated, so as to provide a pop-up window or push notification corresponding to the predicted transaction data to the terminal device, which can be solved by this technical means The problems existing in the prior art, and then achieve the technical effect of improving the user experience.

Although the present invention is disclosed above by the aforementioned embodiments, it is not intended to limit the present invention. Anyone who is familiar with the similar arts can make some changes and modifications without departing from the spirit and scope of the present invention. The scope of patent protection shall be determined by the scope of the patent application attached to this specification.

100: Guidance Systems Based on Predicted Behavior 110: Identity Module 120:Access Mods 130: Behavior record module 140: Classification Module 150:Data module 160: Business Logic Module 170: Learning Mods Step 210: When the terminal device starts to visit the official website of the bank or execute the bank application, identify the user's login identification profile Step 220: Obtain a number plate to correspond to the user's login identification profile, wherein the number plate is random and not reused Step 230: Continuously receive and store the terminal device according to the number plate and the login identification profile since the beginning Visit the bank's official website to leave the bank's official website, self-execute the bank's application to Close the bank application, or visit the bank's official website or execute the bank's application Behavioral data of the application to stay on a single page beyond a preset time Step 240: Continuously receive behavioral data, and classify the behavioral data according to preset conditions to input Travel Behavior Classification Data Step 250: Continuously receive behavior classification data to store as historical classification corresponding to the login identification profile data, and find and output the corresponding historical classifications according to the login identification profile material Step 260: Generate predicted transaction data of the user according to the historical classification data and/or the behavior data, so as to provide a pop-up window or push notification corresponding to the predicted transaction data to the terminal device

FIG. 1 is a system block diagram of an embodiment of the guidance system based on predictive behavior of the present invention. FIG. 2 is a flow chart of an embodiment of a method for implementing the predictive behavior-based guidance system of the predictive behavior-based guidance system of FIG. 1 .

100: Guidance Systems Based on Predicted Behavior

110: Identity Module

120:Access Mods

130: Behavior record module

140: Classification Module

150:Data module

160: Business Logic Module

170: Learning Mods

Claims (6)

A guidance system based on predictive behavior, for when a user operates a terminal device to visit an official bank website or execute a bank application program, the terminal device is provided with a personalized push notification or a pop-up window. The behavior-oriented system includes: an identity recognition module for recognizing a login identification profile (Unique Individual Identifier, UIID) of the user when the terminal device starts to visit the official website of the bank or execute the bank application program; a The access module is used to obtain a number plate (Session ID) to correspond to the user's login identification profile, wherein the number plate is random and not reused; a behavior record module is used to base on the number plate And the login identification profile continues to receive and store the terminal device from the beginning of visiting the bank's official website to leaving the bank's official website, from executing the bank application to closing the bank's application, or from the beginning of visiting the bank's official website or executing the bank's official website. A behavior data that stays on a single page beyond a preset time in the banking application; a classification module is connected to the behavior record module to continuously receive the behavior data, and classify the behavior data according to a preset condition, to output a behavior classification data; a data module, connecting the classification module and the identity recognition module, for continuously receiving the behavior classification data from the classification module to store a history of the paired identification profiles that should be logged in Classification data, and find and output the corresponding historical classification data according to the login identification profile; A business logic module, connected to the data module, for generating a predicted transaction data of the user according to the historical classification data and/or the behavior data, so as to provide the pop-up window or the corresponding predicted transaction data. The push notification is sent to the terminal device, and when the predicted transaction data includes business items that need to be handled over the counter, the business logic module connects to a bank back-end system and is based on the terminal device included in the latest historical classification data According to the location information, find the service locations near the terminal device, the current number of people waiting for each service location, and the certificates and time required to process the business, so that the display content of the push notification or the pop-up window includes the need to visit The business item handled by the counter, the service locations near the terminal device, the current number of people waiting for each service location, the certificates and time required to handle the business, wherein the bank's back-end system is to grasp the address of each service location and the number of people waiting in the current service hall, and a database with the average time spent in handling each business of the bank and the required documents; and a learning module, connecting the behavior record module and the business logic module, for basis The behavior data corresponding to the login identification profile obtains the top-N browsing paths more frequently used by the user, so that the business logic module generates the user according to the top-N browsing paths more frequently used by the user the predicted transaction data, wherein, N is a positive integer greater than zero; wherein, the guidance system based on predicted behavior is also obtained according to the login identification profile of the user and the login identification profile corresponding to the user by the access module the number of such number plates obtained from the number of times the user visited the official website of the bank or executed the bank's application; and In the pop-up window or the displayed content of the push notification, when it is selected to go to the nearby service location to process the predicted transaction data, the online number retrieval service is provided. The predictive behavior-based guidance system according to claim 1, wherein, when the user operates the terminal device to log in to visit the bank's official website or execute the bank application, the identity recognition module uses the user's An account number of the user is used as the login identification profile of the user; when the user operates the terminal device to visit the official website of the bank or execute the bank application program in an unlogged state, the identification module is based on the identification module stored in the terminal device. A browsing information (cookie) identifies the login profile of the user. The guidance system based on predictive behavior according to claim 1, wherein the preset condition includes a transaction target item, and the business logic module counts the historical classification data corresponding to the log-in identification profile according to the transaction target item The number of occurrences of different transaction targets to adjust the weighted index corresponding to each transaction target. When the weighted index corresponding to a certain transaction target exceeds a predetermined value, the content of the pop-up window or the push notification will include the transaction The link to the page corresponding to the target. A guidance method based on predicted behavior is used to provide a personalized push notification or a pop-up window to the terminal device when a user operates a terminal device to visit an official bank website or execute a bank application program, the steps of which include the following steps: : When the terminal device starts to visit the official website of the bank or execute the bank application, identify a login identification profile of the user; obtain a number plate to correspond to the login identification profile of the user, wherein the number plate be random and not reused; Obtain the number of times the user visits the bank's official website or executes the bank's application according to the login identification profile and the number of the number plates obtained for the user's login identification profile; continuous based on the number plate and the login identification profile Receive and store the terminal device from the beginning of visiting the official website of the bank to leaving the official website of the bank, from executing the bank application to closing the bank application, or from the beginning of visiting the official website of the bank or executing the bank application to staying in a single A behavioral data that exceeds a preset time on the page; continue to receive the behavioral data, and classify the behavioral data according to a preset condition to output a behavioral classification data; continue to receive the behavioral classification data to store in pairs and should log in Identifying historical classification data of a profile, and finding and outputting the corresponding historical classification data according to the login identification profile; and generating a predicted transaction data of the user according to the historical classification data and/or the behavior data , to provide the pop-up window or the push notification corresponding to the predicted transaction data to the terminal device, wherein the predicted transaction data of the user is generated according to the historical classification data and/or the behavior data to provide The step of sending the pop-up window or the push notification to the terminal device corresponding to the predicted transaction data includes the following steps: obtaining the top-N browsing paths commonly used by the user according to the behavior data corresponding to the login identification profile, to further base the user's more frequent use The top-N browsing path of the user generates the predicted transaction data of the user, where N is a positive integer greater than zero; when the predicted transaction data includes business items that need to be handled over the counter, connect to a bank back-end system And based on the location information of the terminal device included in the latest historical classification data, find the service locations near the terminal device, the current number of people waiting at each service location, the certificates and time required for handling the business, so that The display content of the push notification or the pop-up window includes the business item that needs to be handled at the counter, the service locations near the terminal device, the current number of people waiting at each service location, the certificates required for the business, and the required time. Among them, the back-end system of the bank is to master the address of each service point and the number of people waiting in the current service hall, and has a database of the average time spent in handling each business of the bank and the required documents; and the pop-up window or the push broadcast If you choose to go to a nearby service location to process the forecast transaction data in the displayed content of the notification, an online number retrieval service will be provided. According to the guidance method based on predicted behavior according to item 4 of the scope of the application, the step of identifying the login identification profile of the user includes: when the user operates the terminal device to access the official website of the bank or executes the execution of the bank in a logged-in state When the application is used, an account of the user is used as the login identification profile of the user; and when the user operates the terminal device to visit the official website of the bank or execute the bank application in an unlogged state, the basis is stored in the A browsing information of the terminal device identifies the login identification profile of the user. The guiding method based on predicted behavior according to item 4 of the scope of application, wherein the preset condition includes a transaction target item, and the guiding method based on predicted behavior further includes the following steps: according to the transaction target item, count the corresponding login identification profile The number of occurrences of different trading targets in the historical classification data is used to adjust the weighted index corresponding to each trading target. When the weighted index corresponding to a certain trading target exceeds a predetermined value, the pop-up window or the push The content of the broadcast notification includes the link to the page corresponding to the transaction target.

Images