CN114926234A - Article information pushing method and device, electronic equipment and computer readable medium - Google Patents

Abstract

The embodiment of the disclosure discloses an article information pushing method and device, electronic equipment and a computer readable medium. One embodiment of the method comprises: generating a user attribute vector and an interest item vector according to a user attribute information group and a user item association point diagram corresponding to each behavior user in a plurality of behavior users to obtain a user attribute vector set and an interest item vector set; clustering the user attribute vector set to generate a user attribute vector cluster group; selecting a user attribute vector cluster corresponding to a target user attribute vector from the user attribute vector cluster group as a target user attribute vector cluster; selecting an interest item vector corresponding to the target user attribute vector cluster from the interest item vector set as an alternative interest item vector to obtain an alternative interest item vector group; and generating a target interest item vector corresponding to the target non-behavioral user according to the candidate interest item vector group. This embodiment improves the accuracy of the items pushed for the new user.

Description

Item information push method, apparatus, electronic device and computer readable medium

technical field

Embodiments of the present disclosure relate to the field of computers, and in particular, to a method, apparatus, electronic device, and computer-readable medium for pushing item information.

Background technique

Item application platforms often recommend items that may be of interest to users through user historical data. At present, the item application platform recommends interesting items for new users, usually by: determining the user type of the new user according to business rules, and determining each item corresponding to the user type as the item that the new user may be interested in.

However, using the above method usually has the following technical problems:

First, because the business rules are artificially set, the items pushed for new users are inaccurate, which wastes the time for new users to browse item information;

Second, the relationship between the new user and the historical users of the item application platform is not considered, so that the pushed items cannot meet the browsing needs of the new user, resulting in a waste of the new user's browsing time.

SUMMARY OF THE INVENTION

This summary of the disclosure serves to introduce concepts in a simplified form that are described in detail in the detailed description that follows. The content section of this disclosure is not intended to identify key features or essential features of the claimed technical solution, nor is it intended to be used to limit the scope of the claimed technical solution.

Some embodiments of the present disclosure propose an item information push method, apparatus, electronic device, and computer-readable medium to solve one or more of the technical problems mentioned in the above background art section.

In a first aspect, some embodiments of the present disclosure provide a method for pushing item information. The method includes: generating a user attribute according to a user attribute information group corresponding to each behavior user among a plurality of behavior users and a user item association node graph vector and interest item vector to obtain the user attribute vector set and the interest item vector set, wherein, the above-mentioned user-item association node graph indicates that there is an association relationship between the behavioral user and the item; the above-mentioned user attribute vector set is clustered to generate User attribute vector cluster group; select the user attribute vector cluster corresponding to the target user attribute vector from the above-mentioned user attribute vector cluster group as the target user attribute vector cluster, wherein, the above-mentioned target user attribute vector is: the user attribute vector corresponding to the target inactive user ; Select the interest item vector corresponding to the above-mentioned target user attribute vector cluster from the above-mentioned interest item vector set as an alternative interest item vector, and obtain an alternative interest item vector group; According to the above-mentioned alternative interest item vector group, generate corresponding to the above target no behavior The target interest item vector of the user, and each item information corresponding to the target interest item vector is pushed to the target terminal.

In a second aspect, some embodiments of the present disclosure provide an apparatus for pushing item information, the apparatus includes: a first generating unit configured to associate a user item with a user attribute information group corresponding to each behavior user among a plurality of behavior users node graph, generating user attribute vector and interest item vector, and obtaining user attribute vector set and interest item vector set, wherein, the above-mentioned user item association node graph indicates that there is an association relationship between behavioral users and items; clustering unit, configured The above-mentioned user attribute vector sets are clustered to generate user attribute vector clusters; the first selection unit is configured to select a user attribute vector cluster corresponding to the target user attribute vector from the above-mentioned user attribute vector cluster groups as the target user Attribute vector cluster, wherein the target user attribute vector is: the user attribute vector corresponding to the target inactive user; the second selection unit is configured to select the interest item vector corresponding to the target user attribute vector cluster from the above-mentioned interest item vector set. As the candidate interest item vector, a candidate interest item vector group is obtained; the second generating unit is configured to generate a target interest item vector corresponding to the above-mentioned target inactive user according to the above-mentioned candidate interest item vector group, and the above target interest Each item information corresponding to the item vector is pushed to the target terminal.

In a third aspect, some embodiments of the present disclosure provide an electronic device, comprising: one or at least one processor; a storage device on which one or at least one program is stored, when one or at least one program is stored by one or at least one The processor executes, so that one or at least one processor implements the method described in any implementation manner of the above first aspect.

In a fourth aspect, some embodiments of the present disclosure provide a computer-readable medium on which a computer program is stored, wherein, when the program is executed by a processor, the method described in any implementation manner of the above-mentioned first aspect is implemented.

The above-mentioned embodiments of the present disclosure have the following beneficial effects: the method for pushing item information in some embodiments of the present disclosure improves the accuracy of items pushed for new users, and avoids wasting time for new users to browse item information. Specifically, the reason for wasting time for new users to browse item information is that the items pushed for new users are inaccurate because the business rules are artificially set, which wastes time for new users to browse item information. Based on this, in the method for pushing item information according to some embodiments of the present disclosure, first, a user attribute vector and an interest item vector are generated according to the user attribute information group and the user item association node graph corresponding to each behavior user among the multiple behavior users, Get user attribute vector set and interest item vector set. Therefore, by generating the user attribute vector set and the interest item vector set, it is convenient to subsequently determine at least one behavioral user that is similar to the user attribute information corresponding to the new user (target inactive user). Next, the above-mentioned user attribute vector set is clustered to generate a user attribute vector cluster group. Therefore, each historical user (behavioral user) can be classified, and the behavioral users with similar user attribute information can be classified into one category, so as to facilitate the subsequent determination of the at least one behavioral user. Next, a user attribute vector cluster corresponding to the target user attribute vector is selected from the user attribute vector cluster group as the target user attribute vector cluster. The above target user attribute vector is: the user attribute vector corresponding to the target inactive user. Thus, each behavioral user similar to the target inactive user can be selected. Then, an interest item vector corresponding to the above-mentioned target user attribute vector cluster is selected from the above-mentioned interest item vector set as a candidate interest item vector, and a candidate interest item vector group is obtained. In this way, the interest item vectors of each behavioral user similar to the target inaction user can be determined, so as to facilitate the determination of the item information of the target inaction user's interest. Finally, according to the candidate interest item vector group, a target interest item vector corresponding to the target inactive user is generated, and each item information corresponding to the target interest item vector is pushed to the target terminal. Therefore, the target interest item vector of the target inactive user can be determined by using the interest item vectors of each behavioral user similar to the target inactive user. Therefore, the pushed item information is more accurate, the accuracy of the item pushed for the new user is improved, and the time for the new user to browse the item information is avoided.

Description of drawings

The above and other features, advantages and aspects of various embodiments of the present disclosure will become more apparent when taken in conjunction with the accompanying drawings and with reference to the following detailed description. Throughout the drawings, the same or similar reference numbers refer to the same or similar elements. It should be understood that the drawings are schematic and that elements and elements are not necessarily drawn to scale.

1 is a flowchart of some embodiments of a method for pushing item information according to the present disclosure;

2 is a schematic diagram of a user item association node graph in the item information push method according to the present disclosure;

3 is a schematic structural diagram of some embodiments of an item information pushing device according to the present disclosure;

4 is a schematic structural diagram of an electronic device suitable for implementing some embodiments of the present disclosure.

Detailed ways

Embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided for a thorough and complete understanding of the present disclosure. It should be understood that the drawings and embodiments of the present disclosure are only for exemplary purposes, and are not intended to limit the protection scope of the present disclosure.

In addition, it should be noted that, for the convenience of description, only the parts related to the related invention are shown in the drawings. The embodiments of this disclosure and features of the embodiments may be combined with each other without conflict.

It should be noted that concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or interdependence.

It should be noted that the modifications of "a" and "at least one" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless the context clearly indicates otherwise, they should be understood as "one or one" at least one".

The names of messages or information exchanged between at least one device in the embodiments of the present disclosure are only for illustrative purposes, and are not intended to limit the scope of these messages or information.

The present disclosure will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

FIG. 1 is a flowchart of some embodiments of methods for pushing item information according to some embodiments of the present disclosure. A process 100 of some embodiments of the method for pushing item information according to the present disclosure is shown. The item information push method includes the following steps:

Step 101: Generate a user attribute vector and an interest item vector according to the user attribute information group and the user item association node graph corresponding to each behavior user among the multiple behavior users, and obtain a user attribute vector set and an interest item vector set.

In some implementations, the executing subject (for example, a computing device) of the item information pushing method may generate the user attribute vector and the interest item according to the user attribute information group and the user item association node graph corresponding to each behavior user among the multiple behavior users vector, get the user attribute vector set and the interest item vector set. The above-mentioned user-item association node graph indicates that there is an association relationship between the behavioral user and the item. The behavioral user among the plurality of behavioral users may be a user who browses item information on the item application platform. The user attribute vector may be a vector representing attribute information of a behavior user. The attribute information of the behavioral user may include, but is not limited to, at least one of the following user attribute values: height information, age information, educational background information, gender information, and region information. The interest item vector may be a vector representing a plurality of items of interest to the behavioral user. The user-item association node graph may include multiple user-item association nodes. The plurality of user item association nodes may include a node corresponding to a user attribute vector and at least one node corresponding to at least one item vector. A connection line is used to connect the node corresponding to the user attribute vector and at least one node corresponding to the at least one item vector. The above-mentioned connecting line may represent that there is an association relationship between the above-mentioned user attribute vector and the item vector. As shown in FIG. 2 , the plurality of user item association nodes may include a node 201 corresponding to the user attribute vector, a node 202 corresponding to the item vector, and a node 203 corresponding to the item vector. The item information corresponding to the node 202 corresponding to the item vector may be a computer. The item information corresponding to the node 203 corresponding to the item vector may be a camera. The node 201 is connected with the node 202 and the node 203 by connecting lines.

In practice, according to the user attribute information group and the user item association node graph corresponding to each behavior user among the multiple behavior users, the above-mentioned execution subject can generate the user attribute vector and the interest item vector through the following steps:

The first step is to normalize the user attribute information that is a continuous feature in the above-mentioned user attribute information group, so as to generate normalized user attribute information, and obtain a normalized user attribute information group. Here, the normalization process may refer to a batch normalization process. Wherein, the user attribute information of the continuous feature may be user attribute information that includes the user attribute value as a continuous numerical value. For example, the user attribute information of the continuous feature may be the height information of the behavioral user.

In the second step, splicing processing is performed on the normalized user attribute information group and the target user attribute information group to generate spliced user attribute information. Wherein, the above-mentioned target user attribute information group is each user attribute information that is a discrete feature in the above-mentioned user attribute information group. The user attribute information of the discrete feature may be user attribute information including user attribute values of discrete numerical values. For example, the user attribute information of the discrete feature may be the user's gender information.

In the third step, vector coding is performed on the above-mentioned spliced user attribute information to generate a spliced user attribute information vector, which is used as a user attribute vector. In practice, the above-mentioned execution body may perform vector encoding processing on the above-mentioned spliced user attribute information by using a vector encoding model, so as to generate a spliced user attribute information vector as a user attribute vector. Here, the vector coding model may be a Bert coding model.

In the fourth step, the above-mentioned user-item association node graph is input into the pre-trained graph neural network to obtain the interest item vector. Here, the graph neural network may be a graph neural network (GCN, Graph Convolutional Network).

Step 102: Perform clustering processing on the user attribute vector set to generate a user attribute vector cluster group.

In some embodiments, the above-mentioned execution body may perform clustering processing on the above-mentioned user attribute vector set by using a k-means clustering algorithm, so as to generate a user attribute vector cluster group. The number of vector clusters corresponding to the above k-means clustering algorithm may be preset. Each user attribute vector in the user attribute vector cluster is relatively similar.

In some optional implementations of some embodiments, the above-mentioned execution body may perform clustering processing on the above-mentioned user attribute vector set through the following steps to generate a user attribute vector cluster group:

The first step is to determine the user attribute set associated with the user attribute vector in the user attribute vector set. Each user attribute represented by the user attribute vector in the user attribute vector set may be determined as a user attribute set. The user attribute vector is generated based on attribute information of each user attribute in the user attribute set. For example, user attributes can be height information, age information, and the like.

In the second step, the number of user attributes included in the above user attribute set is determined as the number of user attributes.

In the third step, the number of each behavior user included in the above-mentioned multiple behavior users is determined as the number of behavior users.

In the fourth step, the number of clusters is determined according to the above-mentioned number of user attributes and the above-mentioned number of behavioral users. First, the product of the above-mentioned number of user attributes and the above-mentioned number of behavioral users may be determined as the total number of user attributes. Then, root-sign processing may be performed on the above-mentioned total number of user attributes, so as to use the total number of user attributes after the root-sign as the number of clusters.

In the fifth step, according to the above-mentioned number of clusters, the above-mentioned user attribute vector set is clustered to generate a user attribute vector cluster group.

In practice, according to the above-mentioned number of clusters, the above-mentioned execution body can perform clustering processing on the above-mentioned user attribute vector set through the following sub-steps, so as to generate a user attribute vector cluster group:

In the first sub-step, the number of clusters of user attribute vectors is randomly selected from the set of user attribute vectors as the initial number of clusters and the number of cluster center vectors.

In the second sub-step, the distance between each user attribute vector and the initial number of cluster center vectors is determined.

The third sub-step is to divide each user attribute vector into the initial user attribute vector cluster corresponding to the cluster center with the closest distance between the initial cluster number and the cluster center vectors, to obtain the initial user attribute vector cluster set.

In the fourth sub-step, a cluster center vector corresponding to each initial user attribute vector cluster in the initial user attribute vector cluster is determined. The cluster center vector is the average vector corresponding to each user attribute vector in the initial user attribute vector cluster.

The fifth sub-step is to determine whether the cluster center vector corresponding to each initial user attribute vector cluster in the initial user attribute vector cluster has changed. That is, whether the cluster center vector corresponding to the initial user attribute vector cluster and the initially determined cluster center vector corresponding to the initial user attribute vector cluster are the same.

A sixth sub-step, in response to determining that no changes have occurred, determine the initial user attribute vector cluster set as the user attribute vector cluster group.

Step 103: Select a user attribute vector cluster corresponding to the target user attribute vector from the user attribute vector cluster group as the target user attribute vector cluster.

In some embodiments, the execution body may select a user attribute vector cluster corresponding to the target user attribute vector from the user attribute vector cluster group as the target user attribute vector cluster. The above target user attribute vector is: the user attribute vector corresponding to the target inactive user. The target inactive user is the inactive user whose corresponding interest item vector is to be determined. An inactive user can be a new user in the item application platform, or a new user viewing an item in the item application platform for the first time.

In practice, the above-mentioned execution body may select the user attribute vector cluster corresponding to the target user attribute vector from the above-mentioned user attribute vector cluster group as the target user attribute vector cluster through the following steps:

The first step is to determine the cluster center vector of each user attribute vector cluster in the above user attribute vector cluster group to obtain a cluster center vector group.

The second step is to determine the vector distance between each cluster center vector in the above-mentioned cluster center vector group and the above-mentioned target user attribute vector to obtain a vector distance group. In practice, the above-mentioned executive body may determine the vector distance between each cluster center vector in the above-mentioned cluster center vector group and the above-mentioned target user attribute vector through a vector distance formula, and obtain a vector distance group. Here, the vector distance formula may be an Euclidean distance formula.

In the third step, the smallest vector distance in the above-mentioned vector distance group is determined as the target vector distance.

In the fourth step, the cluster center vector corresponding to the above target vector distance is determined as the target cluster center vector.

In the fifth step, the user attribute vector cluster corresponding to the center vector of the target cluster is determined as the target user attribute vector cluster.

The relevant content in the optional step 103 is an invention point of the present disclosure, thereby solving the technical problem 2 mentioned in the background art "The association relationship between the new user and the historical user of the item application platform is not considered, resulting in all The pushed items cannot meet the browsing needs of new users, resulting in a waste of new users' browsing time." The factors that cause the waste of new users' browsing time are often as follows: the relationship between new users and historical users of the item application platform is not considered, resulting in the pushed items cannot meet the browsing needs of new users, resulting in a waste of new users' browsing time. If the above factors are solved, the effect of reducing the waste of browsing time of new users can be achieved. First, the cluster center vector of each user attribute vector cluster in the above user attribute vector cluster group is determined to obtain a cluster center vector group. Thus, data support is provided for subsequent selection of behavioral users similar to new users (target inactive users). Next, determine the vector distance between each cluster center vector in the above-mentioned cluster center vector group and the above-mentioned target user attribute vector to obtain a vector distance group. Thereby, it is convenient to select a behavioral user similar to the new user (target inactive user). Next, the smallest vector distance in the above-mentioned vector distance group is determined as the target vector distance. Then, the cluster center vector corresponding to the above target vector distance is determined as the target cluster center vector. From this, the behavioral users most similar to the new user (target inactive user) can be determined. Finally, the user attribute vector cluster corresponding to the center vector of the target cluster is determined as the target user attribute vector cluster. Therefore, it is convenient to use the interested items of the behavioral user similar to the new user (target inactive user) to determine the interested item of the new user. Therefore, it is convenient for the items to be pushed to meet the browsing requirements of the new user, and the waste of browsing time of the new user is reduced.

Step 104: Select the interest item vector corresponding to the target user attribute vector cluster from the above-mentioned interest item vector set as the candidate interest item vector, and obtain the candidate interest item vector group.

In some embodiments, the execution subject may select the interest item vector corresponding to the target user attribute vector cluster from the interest item vector set as the candidate interest item vector to obtain the candidate interest item vector group. In practice, the interest item vector corresponding to each target user attribute vector in the above target user attribute vector cluster may be used as the candidate interest item vector to obtain the candidate interest item vector group.

Step 105: Generate a target interest item vector corresponding to the target inactive user according to the candidate interest item vector group, and push each item information corresponding to the target interest item vector to the target terminal.

In some embodiments, the listing execution body may generate a target interest item vector corresponding to the target inactive user according to the candidate interest item vector group, and push each item information corresponding to the target interest item vector to the target terminal. Here, the target terminal may refer to a terminal to which the account of the target inactive user is logged in.

In practice, the executive body may determine the average value of each candidate interest item vector included in the candidate interest item vector group as the target interest item vector.

In practice, the above-mentioned execution body can push each item information corresponding to the above-mentioned target interest item vector to the target terminal through the following steps:

The first step is to generate item information of each interest item corresponding to the above target inactive user according to the above target interest item vector. In practice, each item information represented by the above-mentioned target interest item vector may be determined as the item information of each interest item corresponding to the above-mentioned target inactive user. In practice, the above-mentioned executive body may also input the above-mentioned target interest item vector into the decoding network to obtain item information of each interest item. The above-mentioned decoding network may be a CNN (Convolutional Neural Network, convolutional neural network).

In the second step, the item information of each of the above-mentioned items of interest is pushed to the above-mentioned target terminal.

The above-mentioned embodiments of the present disclosure have the following beneficial effects: the method for pushing item information in some embodiments of the present disclosure improves the accuracy of items pushed for new users, and avoids wasting time for new users to browse item information. Specifically, the reason for wasting time for new users to browse item information is that the items pushed for new users are inaccurate because the business rules are artificially set, which wastes time for new users to browse item information. Based on this, in the method for pushing item information according to some embodiments of the present disclosure, first, a user attribute vector and an interest item vector are generated according to the user attribute information group and the user item association node graph corresponding to each behavior user among the multiple behavior users, Get user attribute vector set and interest item vector set. Therefore, by generating the user attribute vector set and the interest item vector set, it is convenient to subsequently determine at least one behavioral user that is similar to the user attribute information corresponding to the new user (target inactive user). Next, the above-mentioned user attribute vector set is clustered to generate a user attribute vector cluster group. Therefore, each historical user (behavioral user) can be classified, and the behavioral users with similar user attribute information can be classified into one category, so as to facilitate the subsequent determination of the at least one behavioral user. Next, a user attribute vector cluster corresponding to the target user attribute vector is selected from the user attribute vector cluster group as the target user attribute vector cluster. The above target user attribute vector is: the user attribute vector corresponding to the target inactive user. Thus, each behavioral user similar to the target inactive user can be selected. Then, an interest item vector corresponding to the above-mentioned target user attribute vector cluster is selected from the above-mentioned interest item vector set as a candidate interest item vector, and a candidate interest item vector group is obtained. In this way, the interest item vectors of each behavioral user similar to the target inaction user can be determined, so as to facilitate the determination of the item information of the target inaction user's interest. Finally, according to the candidate interest item vector group, a target interest item vector corresponding to the target inactive user is generated, and each item information corresponding to the target interest item vector is pushed to the target terminal. Therefore, the target interest item vector of the target inactive user can be determined by using the interest item vectors of each behavioral user similar to the target inactive user. Therefore, the pushed item information is more accurate, the accuracy of the item pushed for the new user is improved, and the time for the new user to browse the item information is avoided.

With further reference to FIG. 3 , as an implementation of the methods shown in the above figures, the present disclosure provides some embodiments of an item information push device, these device embodiments correspond to those method embodiments shown in FIG. 1 , the device Specifically, it can be applied to various electronic devices.

As shown in FIG. 3 , the article information pushing apparatus 300 in some embodiments includes: a first generating unit 301 , a clustering unit 302 , a first selecting unit 303 , a second selecting unit 304 and a second generating unit 305 . The first generating unit 301 is configured to generate a user attribute vector and an interest item vector according to the user attribute information group and the user item association node graph corresponding to each behavioral user among the multiple behavioral users, and obtain a user attribute vector set and an item of interest vector set, wherein the above-mentioned user-item association node graph indicates that there is an association relationship between the behavioral user and the item; the clustering unit 302 is configured to perform clustering processing on the above-mentioned user attribute vector set to generate a user attribute vector cluster group; the first selection unit 303 is configured to select the user attribute vector cluster corresponding to the target user attribute vector from the above-mentioned user attribute vector cluster group as the target user attribute vector cluster, wherein, the above-mentioned target user attribute vector is: the target inactive user The corresponding user attribute vector; the second selection unit 304 is configured to select the interest item vector corresponding to the above-mentioned target user attribute vector cluster from the above-mentioned interest item vector set as the candidate interest item vector, and obtain the candidate interest item vector group; The second generating unit 305 is configured to generate a target interest item vector corresponding to the target inactive user according to the candidate interest item vector group, and push each item information corresponding to the target interest item vector to the target terminal.

It can be understood that the units recorded in the apparatus 300 correspond to the respective steps in the method described with reference to FIG. 1 . Therefore, the operations, features, and beneficial effects described above with respect to the method are also applicable to the apparatus 300 and the units included therein, and details are not described herein again.

Referring now to FIG. 4 , a schematic structural diagram of an electronic device 400 suitable for implementing some embodiments of the present disclosure is shown. Electronic devices in some embodiments of the present disclosure may include, but are not limited to, devices such as mobile phones, notebook computers, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Tablet Computers), PMPs (Portable Multimedia Players), and the like Mobile terminals as well as stationary terminals such as digital TVs, desktop computers and the like. The electronic device shown in FIG. 4 is only an example, and should not impose any limitation on the function and scope of use of the embodiments of the present disclosure.

As shown in FIG. 4 , an electronic device 400 may include a processing device (eg, a central processing unit, a graphics processor, etc.) 401 that may be loaded into random access according to a program stored in a read only memory (ROM) 402 or from a storage device 408 Various appropriate actions and processes are executed by the programs in the memory (RAM) 403 . In the RAM 403, various programs and data necessary for the operation of the electronic device 400 are also stored. The processing device 401 , the ROM 402 , and the RAM 403 are connected to each other through a bus 404 . An input/output (I/O) interface 405 is also connected to bus 404 .

Typically, the following devices may be connected to the I/O interface 405: input devices 406 including, for example, a touch screen, touchpad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; including, for example, a liquid crystal display (LCD), speakers, vibration An output device 407 of a computer, etc.; a storage device 408 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 409. Communication means 409 may allow electronic device 400 to communicate wirelessly or by wire with other devices to exchange data. Although FIG. 4 shows electronic device 400 having various means, it should be understood that not all of the illustrated means are required to be implemented or provided. More or fewer devices may alternatively be implemented or provided. Each block shown in FIG. 4 may represent one device, and may also represent at least one device as required.

In particular, according to some embodiments of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, some embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the method illustrated in the flowchart. In some such embodiments, the computer program may be downloaded and installed from the network via the communication device 409 , or from the storage device 408 , or from the ROM 402 . When the computer program is executed by the processing device 401, the above-mentioned functions defined in the methods of some embodiments of the present disclosure are performed.

It should be noted that the computer-readable medium described in some embodiments of the present disclosure may be a computer-readable signal medium or a computer-readable storage medium, or any combination of the above two. The computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus or device, or a combination of any of the above. More specific examples of computer readable storage media may include, but are not limited to, electrical connections having one or at least one wire, portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), erasable Programmable read only memory (EPROM or flash memory), fiber optics, portable compact disk read only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing. In some embodiments of the present disclosure, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. Rather, in some embodiments of the present disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code therein. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device . Program code embodied on a computer readable medium may be transmitted using any suitable medium including, but not limited to, electrical wire, optical fiber cable, RF (radio frequency), etc., or any suitable combination of the foregoing.

In some embodiments, the client and server can communicate using any currently known or future developed network protocol such as HTTP (HyperText Transfer Protocol), and can communicate with digital data in any form or medium (eg, a communications network) interconnected. Examples of communication networks include local area networks ("LAN"), wide area networks ("WAN"), the Internet (eg, the Internet), and peer-to-peer networks (eg, ad hoc peer-to-peer networks), as well as any currently known or future development network of.

The above-mentioned computer-readable medium may be included in the above-mentioned electronic device; or may exist alone without being assembled into the electronic device. The above-mentioned computer-readable medium carries one or at least one program, and when the above-mentioned one or at least one program is executed by the electronic device, the electronic device makes the electronic device: according to the user attribute information group and the user corresponding to each behavior user in the plurality of behavior users Item association node graph, generate user attribute vector and interest item vector, and obtain user attribute vector set and interest item vector set, wherein, the above user item association node graph indicates that there is an association relationship between behavioral users and items; for the above user attributes The vector set is clustered to generate a user attribute vector cluster group; the user attribute vector cluster corresponding to the target user attribute vector is selected from the above-mentioned user attribute vector cluster group as the target user attribute vector cluster, wherein the above-mentioned target user attribute vector is: The user attribute vector corresponding to the target inactive user; select the interest item vector corresponding to the above-mentioned target user attribute vector cluster from the above-mentioned interest item vector set as the candidate interest item vector, and obtain the candidate interest item vector group; vector group, generate a target interest item vector corresponding to the target inactive user, and push each item information corresponding to the target interest item vector to the target terminal.

Computer program code for carrying out operations of some embodiments of the present disclosure may be written in one or more programming languages, including object-oriented programming languages—such as Java, Smalltalk, C++, or a combination thereof, Also included are conventional procedural programming languages - such as the "C" language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer, or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (eg, using an Internet service provider to via Internet connection).

The flowchart and block diagrams in the Figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code that contains one or at least one function for implementing the specified logical function. executable instructions. It should also be noted that, in some alternative implementations, the functions noted in the blocks may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It is also noted that each block of the block diagrams and/or flowchart illustrations, and combinations of blocks in the block diagrams and/or flowchart illustrations, can be implemented in dedicated hardware-based systems that perform the specified functions or operations , or can be implemented in a combination of dedicated hardware and computer instructions.

The units described in some embodiments of the present disclosure may be implemented by means of software, and may also be implemented by means of hardware. The described unit may also be provided in the processor, for example, it may be described as: a processor includes a first generating unit, a clustering unit, a first selecting unit, a second selecting unit and a second generating unit. The names of these units do not constitute a limitation on the unit itself under certain circumstances. For example, the first sending unit may also be described as "in response to detecting the first selection operation acting on the above-mentioned submit control, the above-mentioned The question and answer reply information is sent to the above-mentioned unit of the teacher's terminal".

The functions described herein above may be performed, at least in part, by one or at least one hardware logic component. For example, without limitation, exemplary types of hardware logic components that may be used include: Field Programmable Gate Arrays (FPGAs), Application Specific Integrated Circuits (ASICs), Application Specific Standard Products (ASSPs), Systems on Chips (SOCs), Complex Programmable Logical Devices (CPLDs) and more.

The above descriptions are merely some preferred embodiments of the present disclosure and illustrations of the applied technical principles. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present disclosure is not limited to the technical solution formed by the specific combination of the above-mentioned technical features, and should also cover, without departing from the above-mentioned inventive concept, the above-mentioned Other technical solutions formed by any combination of technical features or their equivalent features. For example, a technical solution is formed by replacing the above-mentioned features with the technical features disclosed in the embodiments of the present disclosure (but not limited to) with similar functions.

Claims (7)

1. A method for pushing item information, comprising: According to the user attribute information group and the user item association node graph corresponding to each behavior user among the multiple behavior users, the user attribute vector and the interest item vector are generated, and the user attribute vector set and the interest item vector set are obtained, wherein the user item The association node graph indicates that there is an association relationship between behavioral users and items; Performing clustering processing on the user attribute vector set to generate user attribute vector cluster groups; The user attribute vector cluster corresponding to the target user attribute vector is selected from the user attribute vector cluster group as the target user attribute vector cluster, wherein the target user attribute vector is: the user attribute vector corresponding to the target inactive user; Select the interest item vector corresponding to the target user attribute vector cluster from the interest item vector set as the candidate interest item vector, and obtain the candidate interest item vector group; According to the candidate interest item vector group, a target interest item vector corresponding to the target inactive user is generated, and each item information corresponding to the target interest item vector is pushed to the target terminal. 2. The method according to claim 1, wherein, generating the user attribute vector and the interest item vector according to the user attribute information group and the user item association node graph corresponding to each behavior user in the plurality of behavior users, comprising: The user item association node graph is input into the pre-trained graph neural network to obtain the interest item vector. 3. The method according to claim 1, wherein the generating a target interest item vector corresponding to the target inactive user according to the candidate interest item vector group, comprising: The average value of each candidate interest item vector included in the candidate interest item vector group is determined as the target interest item vector. 4. The method according to claim 1, wherein the pushing each item information corresponding to the target item of interest vector to the target terminal comprises: generating item information of each interest item corresponding to the target inactive user according to the target interest item vector; The item information of each item of interest is pushed to the target terminal. 5. An item information push device, comprising: The first generating unit is configured to generate the user attribute vector and the interest item vector according to the user attribute information group and the user item association node graph corresponding to each behavior user in the plurality of behavior users, and obtain the user attribute vector set and the interest item vector. set, wherein the user-item association node graph indicates that there is an association relationship between behavioral users and items; a clustering unit, configured to perform clustering processing on the user attribute vector set to generate a user attribute vector cluster group; The first selection unit is configured to select the user attribute vector cluster corresponding to the target user attribute vector from the user attribute vector cluster group as the target user attribute vector cluster, wherein the target user attribute vector is: The user attribute vector of ; The second selection unit is configured to select the interest item vector corresponding to the target user attribute vector cluster from the interest item vector set as the candidate interest item vector, and obtain the candidate interest item vector group; The second generating unit is configured to generate a target interest item vector corresponding to the target inactive user according to the candidate interest item vector group, and push each item information corresponding to the target interest item vector to the target terminal. 6. An electronic device comprising: one or at least one processor; a storage device on which one or at least one program is stored; The one or at least one program, when executed by the one or at least one processor, causes the one or at least one processor to implement a method as claimed in any one of claims 1-4. 7. A computer-readable medium having stored thereon a computer program, wherein the program, when executed by a processor, implements the method of any one of claims 1-4.

Images