CN111092804B

CN111092804B - Information recommendation method, information recommendation device, electronic equipment and storage medium

Info

Publication number: CN111092804B
Application number: CN201911235210.8A
Authority: CN
Inventors: 王伟; 杨海力; 张旭
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2019-12-05
Filing date: 2019-12-05
Publication date: 2022-03-08
Anticipated expiration: 2039-12-05
Also published as: CN111092804A

Abstract

The embodiment of the invention discloses an information recommendation method, an information recommendation device, electronic equipment and a storage medium; the method can obtain a first node and a second node, wherein the first node corresponds to a user in an instant messaging system, the second node corresponds to a user with an information interaction behavior, the first node and the second node are combined to obtain a target network, the target network comprises a target node and an interaction node, the target node corresponds to the target user, the interaction node corresponds to the interaction user with the target user with the information interaction behavior, a propagation influence parameter of each target user is obtained through calculation according to user characteristic information and interaction behavior characteristics of the nodes in the target network, a recommended user is determined from the target users according to the propagation influence parameter of each target user, and recommended information is sent to a terminal corresponding to the recommended user. The embodiment of the invention can improve the accuracy of the result of the propagation influence of the pre-estimated user.

Description

Information recommendation method, information recommendation device, electronic equipment and storage medium

Technical Field

The present invention relates to the field of communications technologies, and in particular, to an information recommendation method, an information recommendation apparatus, an electronic device, and a storage medium.

Background

In some applications, the server may select some users as seed users to promote the product, and the seed users may share the product content through the relationship chain of the social application.

When a seed user is selected, a user with strong propagation capacity is usually selected as the seed user, so that the popularization strength of a product can be improved.

Disclosure of Invention

The embodiment of the invention provides an information recommendation method, an information recommendation device, electronic equipment and a storage medium, which can improve the accuracy of the propagation influence result of a pre-estimated user.

The embodiment of the invention provides an information recommendation method, which comprises the following steps:

acquiring a first node and a second node, wherein the first node corresponds to a user in an instant messaging system, and the second node corresponds to a user with an information interaction behavior in the instant messaging system;

combining the first node and the second node to obtain a target network corresponding to at least one target user, wherein the target network comprises a target node corresponding to the target user and an interactive node, the interactive node corresponds to an interactive user having information interactive behavior with the target user, and the node in the target network has user characteristic information and interactive behavior characteristics;

calculating to obtain a propagation influence parameter of each target user according to the user characteristic information and the interaction behavior characteristics of the nodes in the target network;

determining a recommended user from at least one target user according to the propagation influence parameter of each target user;

and sending recommendation information to a terminal corresponding to the recommendation user.

Correspondingly, an embodiment of the present invention further provides an information recommendation apparatus, including:

the system comprises an acquisition unit and a processing unit, wherein the acquisition unit is used for acquiring a first node and a second node, the first node corresponds to a user in the instant messaging system, and the second node corresponds to a user with an information interaction behavior in the instant messaging system;

the combination unit is used for combining the first node and the second node to obtain a target network corresponding to at least one target user, wherein the target network comprises a target node corresponding to the target user and an interaction node, the interaction node corresponds to an interaction user having an information interaction behavior with the target user, and the node in the target network has user characteristic information and an interaction behavior characteristic;

the calculation unit is used for calculating and obtaining the propagation influence parameters of each target user according to the user characteristic information and the interaction behavior characteristics of the nodes in the target network;

the determining unit is used for determining a recommended user from at least one target user according to the propagation influence parameter of each target user;

and the recommending unit is used for sending recommending information to the terminal corresponding to the recommending user.

Optionally, in some embodiments, the computing unit includes a first sub-fusion unit, a second sub-fusion unit, and a first computing sub-unit,

the first sub-fusion unit may specifically be configured to: fusing the user characteristic information of all nodes in the target network to obtain target user characteristic information;

the second sub-fusion unit may specifically be configured to: fusing the interactive behavior characteristics of all nodes in the target network to obtain target interactive behavior characteristics;

the first computing subunit may be specifically configured to: and calculating the target user characteristic information and the target interaction behavior characteristics of each target node to obtain the propagation influence parameters of each target user.

Optionally, in some embodiments, the system further comprises an extracting unit, an labeling unit and a classifying unit,

the labeling unit may specifically be configured to: extracting feature dimension information of all nodes in the target network;

the labeling unit may specifically be configured to: carrying out feature labeling on the feature dimension information;

the classification unit may specifically be configured to: and classifying according to the marked feature dimension information to obtain the user feature information and the interaction behavior features of all nodes in the target network.

Optionally, in some embodiments, the calculation unit comprises an aggregation unit and a second calculation subunit,

the polymerization unit may be specifically used for: aggregating the target user characteristic information and the target interaction behavior characteristics of the target node to obtain social characteristic information of the target node;

the second calculating subunit may be further specifically configured to: and calculating the social characteristic information of each target node to obtain the propagation influence parameter of each target user.

Optionally, in some embodiments, the aggregation unit further includes a processing unit and an aggregation subunit,

the processing unit may be further specifically configured to: carrying out normalization processing on the target user characteristic information and the target interaction behavior characteristics to obtain a target user characteristic vector and a target interaction behavior characteristic vector;

the polymerization subunit may be further specifically configured to: and aggregating the target user characteristic vector and the target interaction behavior characteristic vector to obtain the social characteristic information of the target node.

Optionally, in some embodiments, the second calculating subunit includes a processing subunit:

the processing subunit may be specifically configured to: and carrying out linear transformation processing on the social characteristic information of each target node to obtain the propagation influence parameter of each target user.

Optionally, in some embodiments, the system further comprises an execution unit,

the execution unit is specifically configured to: acquiring a first node and a second node by adopting an acquisition layer in a preset model, wherein the first node corresponds to a user in the instant messaging system, and the second node corresponds to a user with an information interaction behavior in the instant messaging system;

combining the first node and the second node by adopting a combination layer in a preset model to obtain a target network corresponding to at least one target user, wherein the target network comprises a target node corresponding to the target user and an interaction node, the interaction node corresponds to an interaction user having an information interaction behavior with the target user, and the node in the target network has user characteristic information and an interaction behavior characteristic;

and obtaining the propagation influence parameters of each target user according to the user characteristic information and the interaction behavior characteristics of the nodes in the target network in a preset model calculation layer.

Optionally, in some embodiments, the method further includes:

obtaining a sample first node and a sample second node, wherein the sample first node corresponds to a user in the instant messaging system, and the second node corresponds to the user with information interaction behavior in the instant messaging system;

combining the sample first node and the sample second node to obtain a sample target network corresponding to at least one target user, wherein the sample target network comprises a target node corresponding to the target user and an interaction node, the interaction node corresponds to an interaction user having an information interaction behavior with the target user, and the node in the sample target network has user characteristic information and an interaction behavior characteristic;

obtaining a sample propagation influence parameter of each target user according to the user characteristic information and the interaction behavior characteristics of the nodes in the sample target network;

and converging the sample propagation influence parameter and a preset marking parameter, and returning to the step of obtaining the first node and the second node of the sample until the preset model is trained.

Accordingly, an embodiment of the present invention further provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when executed by the processor, the electronic device implements the steps of any of the methods provided in the embodiments of the present invention.

Correspondingly, the embodiment of the present invention further provides a storage medium, where the storage medium stores instructions, and the instructions, when executed by a processor, implement the steps in any of the methods provided in the embodiments of the present invention.

The embodiment of the invention can obtain a first node and a second node, wherein the first node corresponds to a user in an instant messaging system, the second node corresponds to a user with information interaction behavior in the instant messaging system, then the first node and the second node are combined to obtain a target network corresponding to at least one target user, the target network comprises a target node corresponding to the target user and an interaction node, the interaction node corresponds to an interaction user with information interaction behavior with the target user, the node in the target network has user characteristic information and interaction behavior characteristics, then a propagation influence parameter of each target user is calculated according to the user characteristic information and the interaction behavior characteristics of the node in the target network, a recommended user is determined from at least one target user according to the propagation influence parameter of each target user, and finally, sending recommendation information to a terminal corresponding to the recommended user. According to the method and the device, the propagation influence parameters of the users are estimated by utilizing the user characteristic information and the interactive behavior characteristics of the users, compared with the prior art, the propagation influence of each user in the social network can be estimated more accurately, the recommended users are further determined, and the recommended information is promoted by the recommended users, so that the propagation effect of the recommended information is improved.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

Fig. 1 is a scene schematic diagram of an information recommendation system according to an embodiment of the present invention.

Fig. 2 is a flowchart of an information recommendation method according to an embodiment of the present invention.

Fig. 3 is a schematic view of a deep neural network scene of an information recommendation method according to an embodiment of the present invention.

Fig. 4 is another flow chart of an information recommendation method according to an embodiment of the present invention.

Fig. 5 is a schematic flowchart of another information recommendation method according to an embodiment of the present invention.

Fig. 6 is a schematic view of an application scenario of the information recommendation method according to the embodiment of the present invention.

Fig. 7 is a schematic structural diagram of an information recommendation apparatus according to an embodiment of the present invention.

Fig. 8 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

In the embodiments of the present invention, it is to be understood that terms such as "including" or "having", etc., are intended to indicate the presence of the features, numbers, steps, actions, components, parts, or combinations thereof disclosed in the present specification, and are not intended to exclude the possibility that one or more other features, numbers, steps, actions, components, parts, or combinations thereof may be present or added.

The embodiment of the invention provides an information method, an information recommendation device, electronic equipment and a storage medium.

Referring to fig. 1, an embodiment of the present invention provides an information recommendation system, where the information recommendation system includes a mobile terminal and a server cluster, where the mobile terminal may be a mobile phone, a tablet computer, a notebook computer, and other devices. The server cluster may include at least one of a base server, a virtualized cloud host, and a cloud computing platform (the base server and the virtualized cloud host may further include more, and the specific number is not limited herein), where the base server is a physical Machine, also called a physical server, and is a name of a physical computer relative to a Virtual Machine (Virtual Machine), and the physical Machine provides a hardware environment of the Virtual Machine. By virtualizing the base servers, each base Server can virtualize a plurality of cloud hosts, which are Virtual machines and may also be referred to as Virtual Private Servers (VPSs), and this is a technology for partitioning a Server into a plurality of Virtual independent dedicated servers. The server cluster and the electronic equipment in the information recommendation system can perform collaborative calculation.

The mobile terminal is connected with the server cluster, and an application program supporting social attributes and information recommendation is installed and operated in the mobile terminal. The application may be any of an instant messaging system, a news-pushing system, a shopping system, an online video system, a social-like application that aggregates people based on topics or channels or circles, or other application systems with social attributes. A mobile terminal is a terminal used by a user. Of course, the mobile terminal may include two, three or even more. And the plurality of mobile terminals are connected with the server cluster, and different mobile terminals correspond to terminals used by different users.

Wherein, the server cluster includes: the system comprises an access server and an information recommendation server. The information recommendation server is used for recommending recommendation information (at least one of articles, pictures, audio and video) to the mobile terminal, and the access server is used for providing mobile terminal access service and information recommendation service. The information recommendation server can be one or more. For example, when a preset number of recommended users with higher propagation influence parameters are determined in the candidate user set, the information recommendation server may send recommendation information to a terminal corresponding to the recommended user.

It should be noted that the scene schematic diagram of the information recommendation system shown in fig. 1 is only an example, and the information recommendation system and the scene described in the embodiment of the present application are for more clearly illustrating the technical solution of the embodiment of the present application, and do not form a limitation on the technical solution provided in the embodiment of the present application.

The following are detailed below. The order of the following examples is not intended to limit the preferred order of the examples.

In an embodiment, it will be described from the perspective of an information recommendation device, which may specifically be integrated in a server with processing capabilities.

Referring to fig. 2, an information recommendation method is provided, and a specific process may be as follows:

101. the method comprises the steps of obtaining a first node and a second node, wherein the first node corresponds to a user in an instant messaging system, and the second node corresponds to the user with information interaction behavior in the instant messaging system.

The instant messaging system refers to a system for performing instant messaging between a plurality of users through a client, or a system for performing instant messaging between a plurality of users through a client and a server. Wherein the content of the first and second substances,

the first node represents a corresponding user in the instant messaging system, and a social relationship exists between the first node and the first node. The second node represents a corresponding user with information interaction behavior in the instant messaging system, that is, the second node and the second node have an association relationship of information interaction. For example, user a sends a message to user B and user B receives it, or user B sends a message to user a and user a receives it. Or alternatively. The user A publishes the praise message, the forwarding message, the comment message and the reply message to be visible for the user B, and the praise message, the forwarding message, the comment message and the reply message published by the user B are visible for the user A.

It should be noted that the first node includes at least two, that is, the first node may include two, twenty, two hundred, or even more. The second node comprises at least two, i.e. the second node may comprise two, twenty, two hundred or even more. When the first node and the second node are obtained, the first node and the second node can be randomly obtained from the instant messaging system. Or the first node and the second node are obtained from the same social application program.

For example, a plurality of users are randomly selected from the social application program, the randomly selected users correspond to the first node, the users who have information interaction behavior are randomly selected from the social application program, and the users who have information interaction behavior correspond to the second node.

Of course, in some embodiments, the first node and the second node may be obtained according to a preset rule. For example, the attributes of the users in the social application are classified, and the first node and the second node are obtained from the users with specific attributes. Wherein the attribute may be age, gender, etc. For another example, the users are randomly selected from the users with the age greater than 50 in the social application, the randomly selected users correspond to the first node, the users with the information interaction behavior are randomly selected from the users with the age greater than 50 in the social application, and the users with the information interaction behavior are corresponding to the second node.

Wherein, the propagation influence parameter can also be the number of target nodes propagating influence people.

102. And combining the first node and the second node to obtain a target network corresponding to at least one target user, wherein the target network comprises a target node corresponding to the target user and an interactive node, the interactive node corresponds to an interactive user having information interactive behavior with the target user, and the node in the target network has user characteristic information and interactive behavior characteristics.

It should be noted that the target node is a node corresponding to a target user in the target network. Where the target user may be any one of the target networks. The interactive node is a node having an interactive relationship with the target node.

The user characteristic information can represent information such as attributes, user preferences, living habits, user behaviors and the like of the user. For example, the characteristic information of the user may indicate that the gender of the user is female. The user prefers funny programs. The user's motion is less. Users like online shopping, etc.

The interactive behavior characteristics can represent information such as a chat environment between users, the number of people shared between users and the like. For example, a user chat environment is often on a public information platform, and the number of people leaving a message to the user on the public information platform.

And combining the first node and the second node in a random walk mode to obtain a target network corresponding to at least one target user. The target network is a local network representing a target user. The nodes in the target network comprise target nodes and interactive users having information interaction behaviors with the target nodes. For example, if the user a is the target user, the user a sends a message to the user B and the user B receives the message, or the user B sends a message to the user a and the user a receives the message. And the information interaction action occurs between the user A and the user B, and then the user B is an interactive user. It is understood that the target user connects to a plurality of interactive users who are in information interaction.

It should be noted that the nodes in the target network all have user characteristic information and interactive behavior characteristics.

103. And calculating the propagation influence parameters of the target users according to the user characteristic information and the interactive behavior characteristics of the nodes in the target network to obtain the propagation influence parameters of each target user.

The method comprises the steps of obtaining social characteristic information of a target user according to user characteristic information and interactive behavior characteristics of nodes in a target network, inputting the social characteristic information of the target user into a deep network model, and obtaining propagation influence parameters of the target user through multiple times of linear conversion. And inputting the social characteristic information of each target user into the deep network model to obtain the propagation influence parameter of each target user.

In some embodiments, the calculating, according to the user characteristic information and the interaction behavior characteristics of the nodes in the target network, a propagation influence parameter of the target user to obtain a propagation influence parameter of each target user specifically includes:

(1) fusing the user characteristic information of all nodes in the target network to obtain target user characteristic information;

(2) fusing the interactive behavior characteristics of all nodes in the target network to obtain target interactive behavior characteristics;

(3) and calculating the target user characteristic information and the target interaction behavior characteristics of each target node to obtain the propagation influence parameters of each target user.

The specific process of fusing the user characteristic information of all the nodes in the target network is to coincide the position points of the user characteristic information in each node, and add or matrix-multiply the user characteristic information of the same position point to realize the user characteristic information fusion.

The specific process of fusing the interactive behavior characteristics of all the nodes in the target network is to coincide the position points of the interactive behavior characteristics in each node, and add or matrix-multiply the interactive behavior characteristics of the same position point to realize the interactive behavior characteristic fusion.

The nodes in the target network have multiple nodes, for example, the target network has 5 nodes, each of the 5 nodes includes user characteristic information and interactive behavior characteristics, the user characteristic information of the 5 nodes in the target network is fused to obtain the user characteristic information of the target node, and all the interactive behavior characteristics of the 5 nodes in the target network are fused to obtain the target interactive behavior characteristics.

And inputting the target user characteristic information and the target interaction behavior characteristics of each target node into the deep neural network, and calculating to obtain the propagation influence parameters of each target user.

In some embodiments, before fusing all the user feature information of the nodes in the target network, the method specifically includes the steps of:

(1) extracting feature dimension information of all nodes in the target network;

(2) performing feature labeling on each feature information in the feature dimension information;

(3) and classifying according to the marked feature dimension information to obtain the user feature information and the interaction behavior features of all nodes in the target network.

The nodes in the target network have characteristic dimension information, and the characteristic dimension information is subjected to characteristic extraction to obtain each piece of characteristic information of each node.

And carrying out feature labeling on each feature dimension information. For example, the nodes in the target network include A, B, C, where a1, a2, and A3 are obtained by performing feature labeling on each piece of feature dimension information of a pass, B1, B2, and B3 are obtained by performing feature labeling on each piece of feature dimension information of a pass, and C1, C2, and C3 are obtained by performing feature labeling on each piece of feature dimension information of a pass.

And classifying according to the marked feature dimension information to obtain the user feature information and the interaction behavior features of all nodes in the target network. For example, a1, B1 and C1 are classified into the same class of feature dimension information, a2, B2 and C2 are classified into the same class of feature dimension information, A3, B3 and C3 are classified into the same class of feature dimension information, and a1, B1 and C1 are confirmed to be user feature information, so that a1, B1 and C1 are the user feature information of all nodes in the target network, and a2, B2 and C2 are confirmed to be interactive behavior features, so that a2, B2 and C2 are the interactive behavior features of all nodes in the target network.

In some embodiments, the calculating target user characteristic information and target interaction behavior characteristics of each target node, and calculating propagation influence parameters of target users to obtain the propagation influence parameters of each target user specifically includes the steps of:

(1) aggregating the target user characteristic information and the target interaction behavior characteristics of the target node to obtain social characteristic information of the target node;

(2) and calculating the social characteristic information of each target node to obtain the propagation influence parameter of each target user.

The specific process of aggregating the target user characteristic information and the target interaction behavior characteristics of the target node is as follows: and taking the target user characteristic information and the target interaction behavior characteristics as the input of the image attention network, and inputting the social characteristic information of the target node into the output end of the image attention network after the target user characteristic information and the target interaction behavior characteristics are input into the image attention network.

And calculating social characteristic information of each target node to obtain the propagation influence parameter of each target user. And inputting the social characteristic information of each target node into the deep neural network, so that the propagation influence parameter of each target user can be obtained.

In some embodiments, the aggregating the target user characteristic information and the target interaction behavior characteristic of the target node to obtain the social characteristic information of the target node specifically includes the steps of:

(11) carrying out normalization processing on the target user characteristic information and the target interaction behavior characteristics to obtain a target user characteristic vector and a target interaction behavior characteristic vector;

(12) and aggregating the target user characteristic vector and the target interaction behavior characteristic vector to obtain the social characteristic information of the target node.

And normalizing the target user characteristic information and the target interaction behavior characteristics to obtain a target user characteristic vector and a target interaction behavior characteristic vector, wherein the target user characteristic vector and the target interaction behavior characteristic vector can be used as the input of a graphic attention network.

The target user characteristic vector and the target interaction behavior characteristic vector are used as the input of the graph attention network, the social characteristic information of the target node is obtained after the graph attention network is conducted, and the social characteristic information can be used as the input of the deep neural network.

In some embodiments, the calculating social characteristic information of each target node to obtain propagation influence parameter information of each target user specifically includes:

(1) and carrying out linear transformation processing on the social characteristic information of each target node to obtain the propagation influence parameter of each target user.

The social characteristic information of each target node is input into the deep network model, and the propagation influence parameters of each target user are obtained through multiple linear transformations in the deep network model.

Referring to fig. 3, a specific calculation process is shown, wherein the target user feature information 501 includes a sns vector 501a, an MP vector 501b, a video vector 501c, and the like. The target interaction behavior features 502 include chat room 502a, share title 502b, chat environment 503c, and the like. And social characteristic information 50 aggregating the target user characteristic information 501 and the target interaction behavior characteristic information 502. Inputting the social characteristic information 50 into a relu function in a deep neural network 51, performing cubic linear transformation to obtain a propagation influence probability value of actual output, and calculating the propagation influence parameter of the target user by using a cross entropy function module 53 according to the propagation response probability value of actual output and the propagation influence label 52 expected to be output.

104. And determining a recommended user from at least one target user according to the propagation influence parameter of each target user.

After the propagation influence parameters of each target user are obtained, one or more users with the highest propagation influence can be selected to be determined as recommended users.

In some embodiments, the determining a recommended user from at least one target user according to the propagation influence parameter of each target user specifically includes the steps of:

(1) and calculating the propagation influence parameters of each target node corresponding to the user.

(2) And determining a preset number of users arranged in the front as recommended users.

And inputting the social characteristic information of each target node into the deep network model to obtain the propagation influence parameters of the target user corresponding to each target node. And sequentially sorting according to the level of the propagation influence parameters, and selecting the users in the preset number as recommended users. Wherein the preset number may be determined by a user. For example, the preset number may be 1, 10, 50, or the like. The preset number is not limited in the embodiment of the application.

105. And sending recommendation information to a terminal corresponding to the recommendation user.

Wherein, the recommendation information can be at least one item of articles, pictures, audio and video.

In some embodiments, the propagation influence parameter is obtained by a preset model, including:

1.1, acquiring a first node and a second node by adopting an acquisition layer in a preset model, wherein the first node corresponds to a user in the instant messaging system, and the second node corresponds to the user with information interaction behavior in the instant messaging system.

1.2, combining the first node and the second node by adopting a combination layer in a preset model to obtain a target network corresponding to at least one target user, wherein the target network comprises a target node corresponding to the target user and an interactive node, the interactive node corresponds to an interactive user having information interactive behavior with the target user, and the node in the target network has user characteristic information and interactive behavior characteristics.

And 1.3, calculating the propagation influence parameters of the target users by adopting a calculation layer in a preset model according to the user characteristic information and the interaction behavior characteristics of the nodes in the target network to obtain the propagation influence parameters of each target user.

It should be noted that the method of the above embodiment is not limited to be used in the preset degree model. It may also not be necessary to use in other models.

The step of "training the preset model" may specifically include:

2.1, a sample first node and a sample second node are obtained, wherein the sample first node corresponds to a user in the instant messaging system, and the sample second node corresponds to a user with information interaction behavior in the instant messaging system.

2.2, the sample first node and the sample second node are combined to obtain a sample target network corresponding to at least one target user, the sample target network comprises target nodes corresponding to the target users and interaction nodes, the interaction nodes correspond to interaction users having information interaction behaviors with the target users, and the nodes in the sample target network have user characteristic information and interaction behavior characteristics.

And 2.3, calculating the propagation influence parameters of the target users according to the user characteristic information and the interactive behavior characteristics of the nodes in the sample target network to obtain the sample propagation influence parameters of each target user.

And 2.4, converging the sample propagation influence parameters and preset marking parameters, and returning to the step of obtaining the first node and the second node of the sample until the preset model is trained.

The preset labeling parameter may be obtained through multiple rounds of propagation of the nodes in the second node, for example, the preset labeling parameter is the number of people affected by node propagation. Wherein, the propagation influence parameter can also be the number of target nodes propagating influence people.

For example, a sample first node and a sample second node are obtained as a current training sample, and then nodes in the sample first node and the sample second node are combined to obtain a sample target network, where the sample network includes at least one target node and an interactive user having an information interaction behavior with the target node. And calculating the propagation influence parameters of the target users according to the user characteristic information and the interaction behavior characteristics of the nodes in the sample target network to obtain the social characteristic information of the target users, and calculating the social characteristic information to obtain the sample propagation influence parameters of each target user.

The embodiment of the invention can obtain a first node and a second node, wherein the first node corresponds to a user in an instant messaging system, the second node corresponds to a user with information interaction behavior in the instant messaging system, then the nodes in the first node and the second node are combined to obtain a target network corresponding to at least one target user, the target network comprises a target node corresponding to the target user and an interaction node, the interaction node corresponds to an interaction user with information interaction behavior with the target user, the nodes in the target network have user characteristic information and interaction behavior characteristics, then a propagation influence parameter of each target user is calculated according to the user characteristic information and the interaction behavior characteristics of the nodes in the target network, a recommended user is determined from at least one target user according to the propagation influence parameter of each target user, and finally, sending recommendation information to a terminal corresponding to the recommended user. According to the method and the device, the propagation influence parameters of the users are estimated by utilizing the user characteristic information and the interactive behavior characteristics of the users, compared with the prior art, the propagation influence of each user in the social network can be estimated more accurately, the recommended users are further determined, and the recommended information is promoted by the recommended users, so that the propagation effect of the recommended information is improved.

Referring to fig. 4, another information recommendation method is provided, where the information recommendation detection method may be executed by a server, and a specific process may be as follows:

201. the server acquires a first node and a second node, wherein the first node corresponds to a user in the instant messaging system, and the second node corresponds to the user with information interaction behavior in the instant messaging system.

It should be noted that, an instant messaging system is formed after instant messaging software is installed in the clients of the multiple users, for example, the multiple users download a social application program in an electronic device, the electronic device may be a mobile phone, a tablet computer, a notebook computer, or the like, and the users may be connected through the social application program. The first node represents a corresponding user in the instant messaging system, and a social relationship exists between the first node and the first node. The second node represents a corresponding user with information interaction behavior in the instant messaging system, that is, the second node and the second node have an association relationship of information interaction. For example, user a sends a message to user B and user B receives it, or user B sends a message to user a and user a receives it. Or alternatively. The user A publishes the praise message, the forwarding message, the comment message and the reply message to be visible for the user B, and the praise message, the forwarding message, the comment message and the reply message published by the user B are visible for the user A.

202. The server combines the first node and the second node to obtain a target network corresponding to at least one target user, wherein the target network comprises a target node corresponding to the target user and an interactive node, the interactive node corresponds to an interactive user having information interactive behavior with the target user, and the node in the target network has user characteristic information and interactive behavior characteristics.

It should be noted that the first node and the second node are combined in a random walk manner to obtain a target network corresponding to at least one target user. The target network is a local network representing a target user. The nodes in the target network comprise target nodes and interactive users having information interaction behaviors with the target nodes. For example, in the social application, the user a sends the content of the friend circle and is approved or forwarded by the user B. Then user a is the target user and user B is the interactive user, and correspondingly, user B is the target user and user a is the interactive user.

The user feature information is also the portrait feature of the user, and the user feature information can represent the age, sex, hobby, and the like of the user. The interactive behavior characteristics can represent the chat environment, the number of the chat users and the like between the users.

203. And the server extracts the characteristic dimension information of all the nodes in the target network.

It should be noted that the nodes in the target network have feature dimension information, and the server performs feature extraction on the feature dimension information to obtain each feature information of each node.

204. And the server carries out feature labeling on each feature information in the feature dimension information.

It should be noted that, in the multi-dimensional feature information, feature labeling is performed on each piece of feature information. For example, the nodes in the target network include A, B, C, where a1, a2, and A3 are obtained by performing feature labeling on each piece of feature information of a pass, B1, B2, and B3 are obtained by performing feature labeling on each piece of feature information of B pass, and C1, C2, and C3 are obtained by performing feature labeling on each piece of feature information of C pass.

205. And the server classifies according to the marked feature dimension information to obtain the user feature information and the interactive behavior feature of all nodes in the target network.

It should be noted that, the classification is performed according to the labeled feature dimension information, so as to obtain the user feature information and the interaction behavior features of all nodes in the target network. For example, a1, B1 and C1 are classified into the same class of feature information, a2, B2 and C2 are classified into the same class of feature information, A3, B3 and C3 are classified into the same class of feature information, and a1, B1 and C1 are determined to be user feature information, so that a1, B1 and C1 are the user feature information of all nodes in the target network, and a2, B2 and C2 are determined to be interactive behavior features, so that a2, B2 and C2 are the interactive behavior features of all nodes in the target network.

206. And the server fuses the user characteristic information of all the nodes in the target network to obtain the target user characteristic information.

207. And the server fuses the interactive behavior characteristics of all the nodes in the target network to obtain target interactive behavior characteristics.

208. And the server aggregates the target user characteristic information and the target interaction behavior characteristics of the target node to obtain the social characteristic information of the target node.

It should be noted that, the social characteristic information of the target node is obtained by aggregating the target user characteristic information and the target interaction behavior characteristic of the target node through the graph attention network.

209. And the server linearly transforms the social characteristic information of each target node and calculates to obtain the propagation influence parameter of each target user.

It should be noted that the social characteristic information of each target node is input into the deep network model, and the propagation influence parameters of each target user are obtained through multiple linear transformations in the deep network model. The propagation impact parameter may be a number of propagations affected by the target node. For example, the number of people that the target user can influence through the social application. The number of people affected by the social application may be considered the number of people that have interacted with the target user.

210. And the server determines a recommended user from at least one target user according to the propagation influence parameter of each target user.

It should be noted that, the social characteristic information of each target node is input into the deep network model, so as to obtain the propagation influence parameter of the target user corresponding to each target node. And sequentially sorting according to the level of the propagation influence parameters, and selecting the users in the preset number as recommended users.

211. And the server sends recommendation information to the terminal corresponding to the recommended user.

It should be noted that, articles, pictures, audio, videos, and the like that are closely related to the propagation influence of the recommending user in the social application program are pushed to the terminal of the recommending user.

The embodiment of the invention can obtain a first node and a second node, wherein the first node corresponds to a user in an instant messaging system, the second node corresponds to a user with information interaction behavior in the instant messaging system, then the first node and the second node are combined to obtain a target network corresponding to at least one target user, the target network comprises a target node corresponding to the target user and an interaction node, the interaction node corresponds to an interaction user with information interaction behavior with the target user, the node in the target network has user characteristic information and interaction behavior characteristics, then a propagation influence parameter of each target user is calculated according to the user characteristic information and the interaction behavior characteristics of the node in the target network, a recommended user is determined from at least one target user according to the propagation influence parameter of each target user, and finally, sending recommendation information to a terminal corresponding to the recommended user. The embodiment of the invention can be improved. According to the method and the device, the propagation influence parameters of the users are estimated by utilizing the user characteristic information and the interactive behavior characteristics of the users, compared with the prior art, the propagation influence of each user in the social network can be estimated more accurately, the recommended users are further determined, and the recommended information is promoted by the recommended users, so that the propagation effect of the recommended information is improved.

Referring to fig. 5, in the information recommendation method according to the embodiment of the present application, a first node 601a and a second node 602a are combined in a random walk manner to obtain a target network 603, where the first node 601a is located in a social network 601, and the second node may be a node 602a located in a sharing and clicking network 602, where the social network 601 includes a plurality of first nodes 601a, the first nodes 601a in the social network 601 have a social relationship, the sharing and clicking network 602 includes a plurality of second nodes 602a, and the second nodes 602a in the sharing and clicking network 602 have a sharing and clicking relationship. For example, the user a shares content with the user B, and the user B clicks and views the content shared by the user a, so that the user a and the user B have a relationship of sharing clicks. The target network 603 represents a local area network with a target user, the target network 603 includes a target node 603a and an interactive node 603b, and the interactive node 603b is a node sharing a click behavior with the target node 603 a. The method comprises the steps of fusing user characteristic information of nodes in a target network 603 to obtain target user characteristic information, fusing interaction behavior characteristics of the nodes in the target network to obtain target interaction behavior characteristics, inputting the target user characteristic information and the interaction behavior characteristics into a normalization module 604 for normalization processing to obtain user characteristic vectors and interaction behavior characteristic vectors, and aggregating the user characteristic vectors and the interaction behavior characteristic vectors through a GAT algorithm module 605 to obtain social characteristic information. The social characteristic information is input into the deep neural network 606 to obtain a propagation influence parameter output value, a preset tag propagation number is obtained through multi-round propagation of the shared click network 602, and the preset tag propagation number and the propagation influence parameter output value are subjected to cross entropy calculation, so that the propagation influence parameter of the target user corresponding to the target node is calculated.

According to the method and the device, the propagation influence parameters of the users are estimated by utilizing the user characteristic information and the interactive behavior characteristics of the users, compared with the prior art, the propagation influence of each user in the social network can be estimated more accurately, the recommended users are further determined, and the recommended information is promoted by the recommended users, so that the propagation effect of the recommended information is improved.

Referring to fig. 6, an application scenario diagram of an information recommendation method is further provided in the embodiment of the present application. And determining n users with the strongest sharing and spreading capacity in WeChat social contact as recommended users. The application of the product side is illustrated by a small program of 'watching joy channel' for example, the small program is lighted to a red point in a recommended user, the recommended user is attracted by watching the red point, when other recommended users click the small program related to the red point, the small program jumps to the video related to watching the red point, the video related to watching the red point comprises a plurality of video channels, the video related to watching the red point comprises 'watching joy channels', the contents such as videos or pictures related to the old people are played by 'watching the joy channels', the middle-aged and old people browse and share are mainly triggered by the contents, the earliest seed users are brought, then the maximum user amount is brought to providers through the spreading and sharing of the users in the WeChat, and the size of the watching user is enlarged.

The method described in the above examples is further illustrated in detail below by way of example.

Referring to fig. 7, fig. 7 is a schematic structural diagram of an information recommendation device according to an embodiment of the present application, where the information recommendation device may include an obtaining unit 301, a combining unit 302, a calculating unit 303, a determining unit 304, a recommending unit 305, and the like.

In order to better implement the information recommendation method provided by the embodiment of the present application, the embodiment of the present application further provides a device based on the information recommendation method. The meanings of the nouns are the same as those in the information recommendation method, and specific implementation details can refer to the description in the method embodiment.

The embodiment of the present invention further provides an information recommendation apparatus, including an obtaining unit 301, a combining unit 302, a calculating unit 303, a determining unit 304, and a recommending unit 305: the acquiring unit 301 is configured to acquire a first node and a second node, where the first node corresponds to a user in an instant messaging system, the second node corresponds to a user having an information interaction behavior in the instant messaging system, the combining unit 302 is configured to combine nodes in the first node and the second node to obtain a target network corresponding to at least one target user, the target network includes a target node corresponding to the target user and an interaction node, the interaction node corresponds to an interaction user having an information interaction behavior with the target user, the nodes in the target network have user characteristic information and an interaction behavior characteristic, the calculating unit 303 is configured to calculate a propagation influence parameter of each target user according to the user characteristic information and the interaction behavior characteristic of the nodes in the target network, the determining unit 304 is configured to recommend a user from at least one target user according to the propagation influence parameter of each target user, the recommending unit 305 is configured to send recommendation information to a terminal corresponding to the recommending user.

Optionally, in some embodiments, the calculating unit 303 includes a first sub-fusion unit, a second fusion sub-unit and a first calculating sub-unit,

the first sub-fusion unit may be further specifically configured to: fusing the user characteristic information of all nodes in the target network to obtain target user characteristic information;

the second sub-fusion unit may be further specifically configured to: fusing the interactive behavior characteristics of all nodes in the target network to obtain target interactive behavior characteristics;

the execution unit is specifically configured to: acquiring a first node and a second node by adopting an acquisition layer in a preset model, wherein the first node comprises a plurality of first nodes with user characteristic information, and the first nodes correspond to users in the instant messaging system; the second nodes comprise a plurality of second nodes with interactive behavior characteristics, and the second nodes correspond to users with information interactive behaviors in the instant messaging system;

combining nodes in the first node and the second node by adopting a combination layer in a preset model to obtain a target network corresponding to at least one target user, wherein the target network comprises a target node corresponding to the target user and an interaction node, the interaction node corresponds to an interaction user having information interaction behavior with the target user, and the nodes in the target network have user characteristic information and interaction behavior characteristics;

Optionally, in some embodiments, the method further includes:

obtaining a sample first node and a sample second node, wherein the first node corresponds to a user in the instant messaging system, and the second node corresponds to the user with information interaction behavior in the instant messaging system;

calculating the propagation influence parameters of the target users according to the user characteristic information and the interactive behavior characteristics of the nodes in the sample target network to obtain the sample propagation influence parameters of each target user;

and converging the sample propagation influence parameter and a preset marking parameter, and returning to the step of acquiring the first node and the second node until the preset model is trained.

The obtaining unit 301 in the information recommendation device in the embodiment of the application is configured to obtain a first node and a second node, where the first node includes a plurality of first nodes having user characteristic information, the first node corresponds to a user in the instant messaging system, the second node includes a plurality of second nodes having interactive behavior characteristics, the second nodes correspond to a user having information interactive behavior in the instant messaging system, the combining unit 302 is configured to combine nodes in the first node and the second nodes to obtain a target network corresponding to at least one target user, the target network includes a target node and an interactive node corresponding to the target user, the interactive node corresponds to an interactive user having information interactive behavior with the target user, and nodes in the target network have user characteristic information and interactive behavior characteristics, the calculating unit 303 is configured to calculate a propagation influence parameter of each target user according to the user characteristic information and the interaction behavior characteristics of the node in the target network, the determining unit 304 is configured to determine a recommended user from at least one target user according to the propagation influence parameter of each target user, and the recommending unit 305 is configured to send recommendation information to a terminal corresponding to the recommended user. According to the method and the device, the propagation influence parameters of the users are estimated by utilizing the user characteristic information and the interactive behavior characteristics of the users, compared with the prior art, the propagation influence of each user in the social network can be estimated more accurately, the recommended users are further determined, and the recommended information is promoted by the recommended users, so that the propagation effect of the recommended information is improved.

An electronic device according to an embodiment of the present application is further provided, as shown in fig. 8, which shows a schematic structural diagram of the electronic device according to an embodiment of the present application, specifically:

the electronic device may be a cloud host, and may include components such as a processor 401 of one or more processing cores, a memory 402 of one or more storage media, a power supply 403, and an input unit 404. Those skilled in the art will appreciate that the electronic device configuration shown in fig. 8 does not constitute a limitation of the electronic device and may include more or fewer components than those shown, or some components may be combined, or a different arrangement of components. Wherein:

the processor 401 is a control center of the electronic device, connects various parts of the whole electronic device by various interfaces and lines, performs various functions of the electronic device and processes data by running or executing software programs and/or modules stored in the memory 402 and calling data stored in the memory 402, thereby performing overall monitoring of the electronic device. Optionally, processor 401 may include one or more processing cores; preferably, the processor 401 may integrate an application processor, which mainly handles operating systems, user interfaces, application programs, etc., and a modem processor, which mainly handles wireless communications. It will be appreciated that the modem processor described above may not be integrated into the processor 401.

The memory 402 may be used to store software programs and modules, and the processor 401 executes various functional applications and data processing by operating the software programs and modules stored in the memory 402. The memory 402 may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application program required by at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may store data created according to use of the electronic device, and the like. Further, the memory 402 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid state storage device. Accordingly, the memory 402 may also include a memory controller to provide the processor 401 access to the memory 402.

The electronic device further comprises a power supply 403 for supplying power to the various components, and preferably, the power supply 403 is logically connected to the processor 401 through a power management system, so that functions of managing charging, discharging, and power consumption are realized through the power management system. The power supply 403 may also include any component of one or more dc or ac power sources, recharging systems, power failure detection circuitry, power converters or inverters, power status indicators, and the like.

The electronic device may further include an input unit 404, and the input unit 404 may be used to receive input numeric or character information and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

Although not shown, the electronic device may also include a display processor or the like, which is not described in detail herein. Specifically, in this embodiment, the processor 401 in the electronic device loads the executable file corresponding to the process of one or more application programs into the memory 402 according to the following instructions, and the processor 401 runs the application program stored in the memory 402, thereby implementing various functions as follows:

the embodiment of the invention can obtain a first node and a second node, wherein the first node corresponds to a user in an instant messaging system, the second node corresponds to a user with information interaction behavior in the instant messaging system, then the nodes in the first node and the second node are combined to obtain a target network corresponding to at least one target user, the target network comprises a target node corresponding to the target user and an interaction node, the interaction node corresponds to an interaction user with information interaction behavior with the target user, the nodes in the target network have user characteristic information and interaction behavior characteristics, then the propagation influence parameter of the target user is calculated according to the user characteristic information and the interaction behavior characteristics of the nodes in the target network to obtain the propagation influence parameter of each target user, and then the user is recommended from at least one target user according to the propagation influence parameter of each target user, and finally, sending recommendation information to a terminal corresponding to the recommended user.

In the above embodiments, the descriptions of the embodiments have respective emphasis, and parts that are not described in detail in a certain embodiment may refer to the above detailed description of the information communication method, and are not described herein again.

As can be seen from the above, the electronic device of the embodiment of the present application may obtain a first node and a second node, where the first node corresponds to a user in an instant messaging system, the second node corresponds to a user having an information interaction behavior in the instant messaging system, then combine nodes in the first node and the second node to obtain a target network corresponding to at least one target user, where the target network includes a target node corresponding to the target user and an interaction node, the interaction node corresponds to an interaction user having an information interaction behavior with the target user, the node in the target network has user characteristic information and an interaction behavior characteristic, and then calculate a propagation influence parameter of the target user according to the user characteristic information and the interaction behavior characteristic of the node in the target network to obtain the propagation influence parameter of each target user, and determining a recommended user from at least one target user according to the propagation influence parameters of each target user, and finally sending recommendation information to a terminal corresponding to the recommended user. According to the method and the device, the propagation influence parameters of the users are estimated by utilizing the user characteristic information and the interactive behavior characteristics of the users, compared with the prior art, the propagation influence of each user in the social network can be estimated more accurately, the recommended users are further determined, and the recommended information is promoted by the recommended users, so that the propagation effect of the recommended information is improved.

To this end, embodiments of the present invention provide a storage medium, in which a plurality of instructions are stored, and the instructions can be loaded by a processor to execute the steps in any one of the information recommendation methods provided by the embodiments of the present invention. Such as:

The above operations can be implemented in the foregoing embodiments, and are not described in detail herein.

Wherein the storage medium may include: read Only Memory (ROM), Random Access Memory (RAM), magnetic or optical disks, and the like.

Since the instructions stored in the storage medium can execute the steps in any method provided by the embodiment of the present invention, the beneficial effects that can be achieved by any method provided by the embodiment of the present invention can be achieved, for details, see the foregoing embodiments, and are not described herein again.

The information recommendation method, the information recommendation apparatus, the electronic device and the storage medium provided by the embodiment of the present invention are described in detail above, and a specific example is applied in the present document to explain the principle and the implementation of the present invention, and the description of the above embodiment is only used to help understanding the method and the core idea of the present invention; meanwhile, for those skilled in the art, according to the idea of the present invention, there may be variations in the specific embodiments and the application scope, and in summary, the content of the present specification should not be construed as a limitation to the present invention.

Claims

1. An information recommendation method, comprising:

combining the first node and the second node to obtain a target network corresponding to at least one target user, wherein the target network comprises a target node corresponding to the target user and an interactive node, the interactive node corresponds to an interactive user having an information interactive behavior with the target user, the node in the target network has user characteristic information and an interactive behavior characteristic, and the interactive node comprises nodes which are spread in multiple rounds;

fusing the user characteristic information of all nodes in the target network to obtain target user characteristic information;

fusing the interactive behavior characteristics of all nodes in the target network to obtain target interactive behavior characteristics;

calculating target user characteristic information and target interaction behavior characteristics of each target node to obtain a propagation influence parameter of each target user, wherein the propagation influence parameter is obtained through a preset model, the preset model is obtained by training according to a sample propagation influence parameter and a preset marking parameter, and the preset marking parameter is obtained through multi-round propagation of the nodes;

2. The information recommendation method according to claim 1, wherein before fusing the user feature information of all nodes in the target network, the method comprises:

extracting feature dimension information of all nodes in the target network;

carrying out feature labeling on the feature dimension information;

and classifying according to the marked feature dimension information to obtain the user feature information and the interaction behavior features of all nodes in the target network.

3. The information recommendation method according to claim 1, wherein the calculating target user characteristic information and target interaction behavior characteristics of each target node to obtain a propagation influence parameter of each target user comprises:

aggregating the target user characteristic information and the target interaction behavior characteristics of the target node to obtain social characteristic information of the target node;

and calculating the social characteristic information of each target node to obtain the propagation influence parameter of each target user.

4. The information recommendation method according to claim 3, wherein the aggregating target user characteristic information and target interaction behavior characteristics of the target node to obtain social characteristic information of the target node comprises:

carrying out normalization processing on the target user characteristic information and the target interaction behavior characteristics to obtain a target user characteristic vector and a target interaction behavior characteristic vector;

and aggregating the target user characteristic vector and the target interaction behavior characteristic vector to obtain the social characteristic information of the target node.

5. The information recommendation method according to claim 4, wherein the calculating social feature information of each target node to obtain the propagation influence parameter of each target user comprises:

and carrying out linear transformation processing on the social characteristic information of each target node to obtain the propagation influence parameter of each target user.

6. The information recommendation method according to claim 1, wherein the propagation influence parameter is obtained by a preset model, and comprises:

acquiring a first node and a second node by adopting an acquisition layer in a preset model, wherein the first node corresponds to a user in the instant messaging system, and the second node corresponds to a user with an information interaction behavior in the instant messaging system;

and obtaining the propagation influence parameters of each target user by adopting a calculation layer in a preset model according to the user characteristic information and the interaction behavior characteristics of the nodes in the target network.

7. The information recommendation method according to claim 6, wherein the preset model is trained by:

8. An information recommendation apparatus, comprising:

the combination unit is used for combining the first node and the second node to obtain a target network corresponding to at least one target user, the target network comprises a target node corresponding to the target user and an interaction node, the interaction node corresponds to an interaction user having an information interaction behavior with the target user, the node in the target network has user characteristic information and an interaction behavior characteristic, and the interaction node comprises nodes which are spread in multiple rounds;

the computing unit is used for fusing the user characteristic information of all the nodes in the target network to obtain target user characteristic information; fusing the interactive behavior characteristics of all nodes in the target network to obtain target interactive behavior characteristics; calculating target user characteristic information and target interaction behavior characteristics of each target node to obtain a propagation influence parameter of each target user, wherein the propagation influence parameter is obtained through a preset model, the preset model is obtained by training according to a sample propagation influence parameter and a preset marking parameter, and the preset marking parameter is obtained through multi-round propagation of the nodes;

9. The information recommendation device according to claim 8, further comprising an extraction unit, a labeling unit, and a classification unit,

the labeling unit is used for extracting feature dimension information of all nodes in the target network;

the marking unit is used for marking the characteristic dimension information;

and the classification unit is used for classifying according to the marked feature dimension information to obtain the user feature information and the interaction behavior features of all nodes in the target network.

10. The information recommendation device according to claim 8, wherein the calculation unit further comprises an aggregation unit and a second calculation subunit,

the aggregation unit is used for aggregating the target user characteristic information and the target interaction behavior characteristics of the target node to obtain social characteristic information of the target node;

the second calculating subunit is configured to calculate social characteristic information of each target node, and obtain a propagation influence parameter of each target user.

11. The information recommendation device of claim 10, wherein the aggregation unit further comprises a processing unit and an aggregation subunit,

the processing unit is used for carrying out normalization processing on the target user characteristic information and the target interaction behavior characteristics to obtain a target user characteristic vector and a target interaction behavior characteristic vector;

the aggregation subunit is used for aggregating the target user characteristic vector and the target interaction behavior characteristic vector to obtain social characteristic information of the target node.

12. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the steps of the information recommendation method according to any of claims 1-7 are implemented when the program is executed by the processor.

13. A storage medium having stored thereon a computer program, wherein the computer program when executed by a processor performs the steps of the information recommendation method according to any one of claims 1-7.