HK1136070A - Method for recommending network object information and server thereof - Google Patents

Description

Method and server for recommending network target information to user

Technical Field

The present application relates to the field of computer networks, and in particular, to a method and a server for recommending network target information to a user.

Background

By using the internet technology, the website can recommend various network target information to the user, for example, the website for operating the electronic commerce recommends information about the supply and demand of goods to the user. Because the amount of information that can be provided is quite large and the amount of each user's browsing is limited, websites need to try to allow users to conveniently find the desired information.

For this, it is now common practice to: recording the information registered by the user in the website, or presuming the information possibly needed by the user according to the location of the IP address of the network protocol of the user; this information is provided to the user when the user visits the website. Taking a website for operating e-commerce as an example, the recorded information registered by the user on the website includes the industry to which the user belongs, income interval, interested commodity categories and personal hobbies of the user, and the information of which commodities the user specifically needs can be inferred according to the information. In addition, according to the location of the IP address of the user, the commodity information which is possibly needed by the user can be presumed according to the geographical characteristics of the location.

In the actual operation of the website, the above method has certain limitations. First, the information filled in by the user during registration is not necessarily in accordance with the actual situation or is very incomplete, and some of the registration information filled by the user may change continuously. For example, the income interval of the user may change continuously, but in most cases, the user will not continuously modify such personal information. Second, the products of interest may change over time, such as with a change in work units, or with a change in procurement tasks within the same unit. Neither of these information can be clearly described at the time of registration. Finally, the commodity information which may be needed by the user is presumed according to the IP address location of the user, the commodity range determined by the method is still quite fuzzy, and the actual location of the user is not consistent with the IP address displayed by the user under the condition that the user uses the foreign agent server to connect to the Internet. Thus, according to current practice, it remains difficult to efficiently provide users with network target information that they may need.

Disclosure of Invention

The main purpose of the present application is to provide a method and a server for recommending network target information to a user, so as to solve the problem in the prior art that it is difficult to effectively provide network target information that may be needed by the user to the user.

In order to solve the above problems, the present application provides the following technical solutions:

a method of recommending network objective information to a user, comprising:

the server determines the times of executing the network behaviors on each network target by the user in a set time period according to the preselected network behaviors and the network targets;

the server selects a network target according to the determined times;

the server provides the user with information of the selected network object.

A server for recommending network objective information to a user, comprising:

the determining module is used for determining the times of executing the network behaviors on each network target by the user in a set time period according to the preselected network behaviors;

the selection module is used for selecting the network target according to the times determined by the determination module;

and the providing module is used for providing the information of the network target selected by the selecting module for the user.

A server for recommending network objective information to a user, comprising:

the database module is used for storing network target information;

the determining module is used for determining the times of executing the network behaviors on each network target by a user in a set time period according to the preselected network behaviors and the network targets;

the selection module is used for selecting the network target according to the times determined by the determination module;

and the providing module is used for extracting the information of the network target selected by the selecting module from the database module and then providing the information to the user.

According to the technical scheme of the embodiment of the application, the times of the specific network behaviors of the user to the network target in a certain time period are counted, the degree of interest of the user to the network target information is analyzed according to the times, and the network target information is provided for the user according to the degree of interest. Because the specific network behavior of the user on the network target can directly and truly reflect the degree of the requirement of the user on the network target information, according to the scheme of the embodiment of the application, the information of which network targets are really needed by the user can be accurately obtained, and the network target information needed by the user can be effectively provided for the user.

Drawings

Fig. 1 is a schematic diagram of an internet structure in an embodiment of the present application;

FIG. 2 is a flow chart of a method in an embodiment of the present application;

FIG. 3(a) is a schematic diagram of classifying and storing user network behavior data in an embodiment of the present application;

FIG. 3(b) is a schematic diagram illustrating recommendation of network target information to a plurality of users in an embodiment of the present application;

FIG. 4(a) is a schematic structural diagram of a server in an embodiment of the present application;

FIG. 4(b) is a schematic diagram of another server structure in the embodiment of the present application;

FIG. 5 is a schematic structural diagram of a selection module of a server in an embodiment of the present application;

fig. 6 is another schematic structural diagram of a selection module of a server in the embodiment of the present application.

Detailed Description

The technical solutions of the embodiments of the present application are described below with reference to the accompanying drawings, and the embodiments of the present application are not limited to the forms of the drawings in various implementations.

As shown in the internet structure of fig. 1, a user accesses a server 12 through a terminal device 11 to acquire information of a network object. There is typically more than one user in the internet accessing a server as shown in fig. 1. In order to effectively provide the user with the information of the network target required by the user, in the embodiment of the present application, the server 12 recommends the information of the network target to the user according to the times of various network behaviors actually made by the user on the network target. The information of the network object may be stored in the database means 13. In the internet, the database device 13 may also be provided in the server 12 as a component of the server. When the server 12 recommends the information of the network target to the user, the process may be performed according to the flow shown in fig. 2, and the specific steps are as follows:

step 21: and determining the times of executing the network behaviors on each network target by the user in a set time period according to the preselected network behaviors and the network targets.

Step 22: the network object is selected based on the counted number of times in step 21.

Step 23: the user is provided with information of the network object selected in step 22.

The technical solutions in the embodiments of the present application are further described below.

The commodities on the Internet e-commerce website all have the product description keywords of the commodities. For example, a certain enterprise publishes "R-brand rice" on a website, and according to the website requirements, the enterprise fills in the description keywords "R-brand", "rice", "northeast", "long grain", "glutinous rice" and the like of the product at the time of publication.

When accessing the internet, users may take various network actions, such as: the commodity information is distributed via the internet, an electronic mail containing the commodity information is received via the internet, the commodity information is browsed via the internet, the commodity information is searched via the internet, information for evaluation of the commodity is distributed via the internet, and the like. Here, information on evaluation of a product is distributed via the internet, for example, a score is given to the product, the product is recommended, or a character giving a comment on the product is given. The network side stores information about the network behavior of the user, usually in a log of the website, or in a database or other location of the website. This information is growing continuously during network operation and can therefore be handled in categories, with new information resulting from network behavior being stored in categories each time a user performs the network behavior. Classification may be performed by user, and as shown in fig. 3(a), the network behavior information of each existing user and the network behavior information of the newly added user are stored by user classification, so that the data required in step 21 may be extracted from the data stored by classification. When the information brought by the user network behavior is classified according to the user, the information can be further classified according to various network behaviors of the user, for example, a user performs 2 searching behaviors and 3 publishing behaviors, and receives 4 pieces of advertisement mails related to commodities, then the information of the 2 searching behaviors, such as keywords of searching and searching time, can be recorded in the "searching behavior" class of the user, and similarly, the information of the 3 publishing behaviors and the actions of receiving 4 pieces of advertisement mails related to commodities can be respectively recorded in the "publishing behavior" and the "subscribing behavior" of the user. In this way, when the database stores various types of behavior information of each user, it is convenient to count the information items, i.e., the user network behaviors, at the same time, and as will be seen in the following description, various types of behaviors of the user have different weights in the statistics in step 22, so storing various types of behaviors of the user in a classified manner in advance is helpful to improve the efficiency of executing step 21 and step 22. Step 21 mainly comprises the following steps:

step 211: and extracting the stored user network behavior information. If the network side stores some user network behavior information which is not classified according to users, the information can be extracted first.

Step 212: and receiving the network behavior information of the newly added user. In an implementation, step 212 is an ongoing process.

Step 213: and classifying and storing the data obtained in the steps 211 and 212.

Step 214: for the current user, the data of the network behavior of the user is extracted from the data saved in step 213. As will be seen in the following analysis, the information that is extracted about the behavior of the user's network in this step may be selectively extracted.

Step 215: the number of times the current user performs network actions on each network target is determined based on the data extracted in step 214.

The steps 211-213 are the preparation before data statistics. Statistical behavior is implemented in steps 214 and 215. This is further analyzed below for ease of understanding. If a user is interested in a network object, he may generally perform some network behavior on the network object, such as searching the network object, subscribing to e-mails about the network object, making comments on the network object, etc. When accessing the internet, a user can take various network actions, which can be regarded as being performed aiming at various network targets in the internet. For example, if a user obtains information of a commodity R-brand rice through the internet, the user may be considered to perform an action of obtaining the information on a network target "R-brand rice", or may be considered to perform the action on a network target "R-brand grain product", "rice", or "grain product". When a user conducts a commodity transaction or other activities related to commodities on the internet, the network object is generally a specific commodity of a specific brand, such as "R-brand rice" herein, and may further include information such as the model of the commodity. Because the embodiment of the application considers how to judge the network target information interested by a user through the behavior of the user, according to a specific behavior of the user, namely 'obtaining information of the commodity R-brand rice', the trend of the user to the network target shown by the behavior can be analyzed from multiple angles, for example, the user can be presumed to like eating the R-brand rice, or the user can be presumed to accept the R-brand food product, and the user can be presumed to need to purchase the rice at present. It can be seen that the reason why the possibility of multiple possibilities can be inferred is that the network object generally has multiple attributes, such as its brand or its variety, and the network object can be classified into various categories according to the various attributes, and for goods transacted via the internet, it is generally classified into the category of the goods. In the conventional device for storing commodity information in the internet, commodity categories are generally adopted to manage the commodity information. The category of the product is a collection name to which the product belongs after the product is classified according to a certain rule. The rules can be divided into a plurality of ways, for example, the electronic products are divided according to brands, the brands of the electronic products are A cards, B cards, C cards and the like, the categories of the electronic goods are categories 1, 2 and 3, and the members of the three categories respectively comprise the electronic products of the A cards, the B cards and the C cards. The commodities can also be divided according to functions, for example, the commodities are divided into a printer, a digital camera and a mobile phone, three categories can be correspondingly set, and the information of the three categories of commodities can be respectively stored. In the embodiment of the application, the R brand rice can be classified into the categories of 'R brand grain products' and 'rice'. The network behavior performed by a user on a network object typically reflects his interest in a certain type of purpose to which the network object belongs. The category of the goods can be targeted as a network. Sometimes, keywords are also used to divide various commodities, so that a specific commodity can correspond to a plurality of keywords, for example, "R brand rice" can correspond to keywords "R brand" and "rice", and each keyword actually marks a category, for example, the keyword "R brand" marks R brand agricultural products, and the keyword "rice" marks all brands of rice. These keywords can also be used as network targets to examine the user's preference for the keywords, thereby selectively providing the user with network information containing the keywords of his preference. The user may perform a search action among network actions when obtaining information, i.e., a search using keywords. For example, entering the keyword "R-brand rice" for searching, here actually two keywords "R-brand" and "rice", such a query may be considered a one-time search action for the two keywords. Generally, when a user executes a search behavior, an internet page containing a form is acquired through terminal equipment, a keyword is input in the form and then submitted to a server on a network side, and the server acquires a corresponding keyword according to the form submitted by the user. There may be mutual association between the network behaviors of the users, for example, the users click and browse the search results after the search behavior, as in the above example, the users provide the keyword "R-brand rice" to search, after obtaining the search result about R-brand rice, the users will then click the link in the search result, so as to browse various commodities, at this time, the description keywords (description keywords are generally provided by the user who issued the commodity when the commodity is issued and stored on the network side, for example, the user issues information about a kind of rice, the description keywords of such rice are filled in: R-brand, rice, sticky rice, northeast, etc.) that the commodities that the users further browse after searching the keyword "R-brand rice" have description keywords "sticky rice", "northeast", etc., if the users browse more information describing that the keywords are sticky rice, the description keyword "glutinous rice" may be targeted for a network to preferentially provide the user with information of glutinous rice regardless of the brand of glutinous rice.

In the embodiment of the present application, a value is used to measure the preference or interest of the user to the network target, and the value is obtained by counting the times of performing the network behavior on the network target by the user. In this embodiment, when counting the number of times of the network behavior of the user, the number of times of the user performing the network behavior on a certain network target in a time period is counted, one or more network behaviors are specified, and a plurality of network targets are specified for counting. For example, consider the following network behaviors of a certain user a on the network target "R-brand rice": receiving an e-mail containing the information of the ' R-brand rice ' through the internet, searching the information of the ' R-brand rice ' through the internet, and issuing the information of the evaluation of the ' R-brand rice ' through the internet, and analyzing the degree of the user's preference for the ' R-brand rice ' according to the behaviors, counting can be performed from a selected starting time, namely, step 215 is performed, and the number x1 of searches of the user a for the ' R-brand rice ' from zero point of 1 day of 3 months to zero point of the next day, the number y1 of received mails containing the information of the ' R-brand rice ', and the number z1 of the user a issuing the evaluation of the ' R-brand rice ' are counted. According to step 214, the data required for the statistics is obtained. Similarly, the above three network behavior times of 3 months and 2 days are counted and recorded as x2, y2 and z2, respectively. These numbers of times of the multi-day statistics are added to obtain a value, for example, the value obtained by statistics from day 1 to day 31 of 3 months is denoted as R (x, y, z, d (1, 31)), that is:

R(x，y，z，d(1，31))＝[x1+x2+...+x31]+[y1+y2+...+y31]+[z1+z2+...+z31]。

the value R (x, y, z, d (1, 31)) may reflect to some extent the preference of the user a for "R-brand rice". If the preference degree and the preference of the user A to the commodity are analyzed, not only the network behavior value of the user A related to the 'R-brand rice' but also the network behavior value of the user A to other network targets, namely other brands such as the S-brand rice and the T-brand rice, are calculated. After the times of the network behaviors of the user A on the rice of each brand are obtained through statistics, the times are sorted according to the step 22 and the step 23, a plurality of brands corresponding to larger times are selected from the times, and the information of the brands is provided for the user A.

In the above example, the category of each brand mark is used as the network target, and another group of network targets may be determined according to the keyword for statistics, for example, the keywords "glutinous rice", "northeast rice" and "long grain" are used as the network targets, and all network behaviors (information distribution, browsing, searching, mail notification, etc.) of the user a are summarized and counted according to the above method to obtain a value R (glutinous rice, user a) representing the number of times of all network behaviors of the user a on "glutinous rice"; similarly, the numerical values of R (northeast rice, user A), R (long grain, user A) and the like are calculated, and if the numerical value of R (glutinous rice, user A) is the maximum, the user A has the maximum preference degree on glutinous rice recently. A rice commodity to which glutinous rice is recommended accordingly.

Generally, the following formula (1) is given:

cnt (Act (i), Time (j), K)₁) Indicates that in the j time period Time (j), the network target K is₁The number of times the i-th network action Act (i) is executed, and the value C (K)₁) The user A can reflect the network target K to a certain extent₁The degree of interest. For other network targets, for example, various rice brands are used as a plurality of network targets K₂、K₃、......、K_nThe corresponding value C (K) can also be calculated₁)、C(K₂)、......、C(K_n) Selecting information of network target according to magnitude relation of the values and providing the information to user, for example, C (K)₁)、C(K₂)、......、C(K_n) And sequencing, namely the network targets corresponding to the top numerical values are concerned by the user, so that the information containing the network targets, namely the rice brands, can be preferentially displayed to the user when the user browses the network. The network target can also be various keywords of commodity connotation, such as glutinous rice, northeast, long grains, polished round-grained rice and the like, which are used as the network target, the corresponding C value is calculated according to the network behavior data of the user A, and if the two keywords are the glutinous rice and the northeastThe corresponding C value is arranged in the first two digits, the two keywords can be further matched to form description about the network object, namely the northeast glutinous rice, so that the user A can be considered to have higher preference degree on the northeast glutinous rice, and glutinous rice information produced in the northeast can be displayed on a homepage seen by the user A after the user A logs in.

In practice, due to the limitation of the number of commodities, the number may not be enough when the user may be recommended to include the keyword "glutinous rice, northeast, long grains". Therefore, the category of the commodity in various network behaviors of the user can be used as a network target, and the C value can be calculated.

For example, the user browses the R-brand rice, takes the category 'rice' as a network target, counts the network behaviors of the user about the 'rice', and calculates a corresponding C value.

Generally, the following formula (1) is given:

in this way, it is calculated which categories a user is most interested in. When the commodity recommendation is made to the user on the website, the categories and the keywords can be combined according to the quantity of the commodities, the requirements of the website and the like to make the commodity recommendation. If the calculated numerical value of the preference degree of the user A to the rice of various brands is generally higher, the user mainly purchases the rice; or if the value is not high and the value reflecting the user's preference for other agricultural products is high, it means that purchasing rice is only one of the user's actions. Therefore, the method in the embodiment of the application can determine which network targets are interested by the user more specifically.

In addition, as can be seen from the above method, the more the number of dates counted, the more the added value can truly reflect the user's interest level in the network target. According to the scheme of the embodiment of the application, the statistical date can be as long as one year. But if C (K) is calculated from a long past date₁) A value, the user's interest may have shifted and the persuasion of the value may be diminished. In addition, the network behavior can reflect the attention degree of the user to the network target. For example, a user typically makes a comment on a product to indicate that he is interested in the product. The user obtains information of a certain commodity, which may be accidental behavior, and does not necessarily indicate that he is concerned about the commodity. To take into account the above analysis, C (K) is calculated₁) In this case, the weighted value may be assigned to the number of times of various network behaviors performed on the network object by the user in each time period, and then the weighted sum of the number of times for a plurality of time periods may be calculated. The weight value can be distributed according to the network behavior, the occurrence time of the network behavior or both. If the weight of the ith network behavior is represented as twc (act (i)), and the weight value of the ith network behavior act (i) executed in the jth time period time (j) is represented as tdf (time (j)), C (K) may be calculated according to the following equation₁) The value:

where n represents the number of network actions considered and T represents the number of time periods considered. Of course, twc (act (i)) and tdf (time (j)) may be counted as one of the items. The farther a time period is from the current time, the lower the weight value corresponding to the time period should generally be. For the relationship between time and weight, the same principle should be applied to each commodity. A positive number less than 1 may be used as the weight for each time segment, e.g., a time segment 1 year ago, with a weight of 0.1; the time periods of the first four months, the fifth to eighth months, and the latter four months within 1 year may be weighted to 0.2, 0.3, and 0.4, respectively. Of course, other positive numbers may be selected as the weight, and in general, the weight may be smaller in a time period with a longer time.

For other network targets K₂、K₃、......、K_nThe corresponding value C (K) can also be calculated₁)、C(K₂)、......、C(K_n) Similarly to the previous method, based on the respective weighted sums C (K) calculated₁)、C(K₂)、......、C(K_n) Providing the user with information of the selected weighted sum corresponding to the network object. Similar processing can be done for each user. As shown in fig. 3(b), data generated by the network behavior of each user is saved, the data is analyzed according to the method to obtain the network targets preferred by each user, and finally, the information of the network targets is provided for the user.

Based on the method in the embodiment of the present application, a server in the embodiment of the present application is explained below. In the embodiment of the present application, the modules in the server are divided according to their functions, and may be implemented by using software, or may be implemented by using hardware or a combination of hardware and software, where the software may be stored in a storage device in the form of a magnetic disk, an optical disk, or an integrated circuit. As shown in fig. 4(a), the server 40a is used for providing the network target information to the user, and includes a determining module 41, a selecting module 42 and a providing module 43. The determining module 41 is configured to determine, according to the preselected network behavior, the number of times that the user performs the network behavior on each network object in the set time period. The selecting module 42 is configured to select the network target according to the times determined by the determining module 41. The providing module 43 is used for providing the user with information of the network object selected by the selecting module 42.

A database module may be further provided in the server 40a, as shown in fig. 4(b), and the server 40b includes a determination module 41, a selection module 42, and a providing module 43, and further includes a database module 44 for storing the network object information. Thus, when the providing module 43 provides the network object information to the user, the information of the network object selected by the selecting module 42 is extracted from the database module 44, and then the extracted information is provided to the user.

One configuration of the selection module 42 is shown in fig. 5, and includes a calculation unit 451 for calculating, for each network objective, a sum of the number of times that the network objective is performed by the user in a selected plurality of time periods; and a selection unit 452 configured to select a network destination according to the sum of the times calculated by the calculation unit 451.

Another structure of the selecting module 42 is shown in fig. 6, and includes a calculating unit 461, configured to, for each network target, assign a preset weight value to the number of times of each network action performed on the network target by the user in each of the time periods, and then calculate a weighted sum of the number of times of the network actions performed on the network target by the user in the selected time periods; and a selecting unit 462, configured to select a network target according to the weighted sum calculated by the calculating unit 461.

According to the technical scheme of the embodiment of the application, the times of the specific network behaviors of the user to the network target in a certain time period are counted, the degree of interest of the user to the network target information is analyzed according to the times, and the network target information is provided for the user according to the degree of interest. Because the specific network behavior of the user on the network target can directly and truly reflect the degree of the requirement of the user on the network target information, according to the scheme of the embodiment of the application, the information of which network targets are really needed by the user can be accurately obtained, and the network target information needed by the user can be effectively provided for the user. And this way can promote the efficiency of providing information to users by the internet and enable users to conveniently obtain information needed by themselves from the internet.

It will be apparent to those skilled in the art that various changes and modifications may be made in the present application without departing from the spirit and scope of the application. Thus, if such modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is intended to include such modifications and variations as well.

Claims (11)

1. A method for recommending network target information to a user, comprising:

the server determines the times of executing the network behaviors on each network target by the user in a set time period according to the preselected network behaviors and the network targets;

the server selects a network target according to the determined times;

the server provides the user with information of the selected network object.

2. The method of claim 1, wherein the network objective comprises: the category of the commodity traded through the Internet;

or comprises the following steps: and keywords corresponding to the commodity transacted through the Internet.

3. The method of claim 2, wherein the network behavior comprises one or more of:

releasing commodity information through the Internet;

receiving an e-mail containing commodity information through the internet;

issuing information for commodity evaluation through the Internet;

acquiring commodity information through the Internet;

commodity information is searched through the internet.

4. The method of claim 1, wherein the server selecting a network target according to the determined number of times comprises:

calculating the sum of the times of executing network behaviors on the network target by the user in a plurality of selected time periods aiming at each network target;

selecting from among a plurality of network objectives according to the sum of the times calculated for the plurality of network objectives.

5. The method of claim 1, wherein selecting a network target according to the determined number of times comprises:

for each network target, distributing a preset weight value to the times of each network behavior executed on the network target by the user in each time period, and then calculating the weighted sum of the times of the network behaviors executed on the network target by the user in a plurality of selected time periods;

selecting from among a plurality of network objectives based on the weighted sum computed for the plurality of network objectives.

6. The method of claim 5, wherein the weight comprises a weight value assigned according to a type of the network behavior, a weight value assigned according to an occurrence time of the network behavior, or a product of the two.

7. The method of claim 5, wherein calculating the weighted sum of the number of times the user performed the network action on the network object over the selected plurality of time periods comprises calculating according to the following equation:

wherein, P (K)₁) Denotes the weighted sum, Cnt (Act (i), Time (j), K₁) Indicates that in the j time period Time (j), the network target K is₁The number of times the ith network behavior act (act) (i) is executed, twc (act (i)) indicates a weight value of the ith network behavior act (i), tdf (time (j)) indicates a weight value of the ith network behavior act (i) executed in the jth time period time (j), n indicates the number of considered network behaviors, and T indicates the number of considered time periods.

8. A server for recommending network target information to a user, comprising:

the determining module is used for determining the times of executing the network behaviors on each network target by a user in a set time period according to the preselected network behaviors and the network targets;

the selection module is used for selecting the network target according to the times determined by the determination module;

and the providing module is used for providing the information of the network target selected by the selecting module for the user.

9. The server according to claim 8, wherein the selection module comprises:

the calculating unit is used for calculating the sum of the times of executing the network behaviors on the network target by the user in a plurality of selected time periods aiming at each network target;

and the selection unit is used for selecting the network target according to the sum of the times calculated by the calculation unit.

10. The server according to claim 8, wherein the selection module comprises:

the calculation unit is used for distributing a preset weight value to the times of each network behavior executed on the network target by the user in each time period aiming at each network target, and then calculating the weighted sum of the times of the network behaviors executed on the network target by the user in a plurality of selected time periods;

and the selecting unit is used for selecting the network target according to the weighted sum calculated by the calculating unit.

11. A server for recommending network target information to a user, comprising:

the database module is used for storing network target information;

the determining module is used for determining the times of executing the network behaviors on each network target by the user in a set time period according to the preselected network behaviors;

the selection module is used for selecting the network target according to the times determined by the determination module;

and the providing module is used for extracting the information of the network target selected by the selecting module from the database module and then providing the information to the user.