CN108416645B

CN108416645B - Recommendation method, device, storage medium and equipment for user

Info

Publication number: CN108416645B
Application number: CN201810054377.3A
Authority: CN
Inventors: 张勇
Original assignee: Beijing Sankuai Online Technology Co Ltd
Current assignee: Beijing Sankuai Online Technology Co Ltd
Priority date: 2018-01-19
Filing date: 2018-01-19
Publication date: 2021-08-20
Anticipated expiration: 2038-01-19
Also published as: CN108416645A

Abstract

The embodiment of the invention provides a recommendation method, a recommendation device, a storage medium and computer equipment for a user. The method comprises the steps of establishing an association relationship between every two users according to the characteristic information of the users, establishing at least one user group among all the users, determining the characteristic users in the user group, recommending the network object based on the characteristic users, dividing the users into different groups according to the characteristic information of the users, enabling the users with the association relationship to have certain similarity to the preference of the network object, recommending service aiming at the user group, recommending the service or commodity which the users really have requirements, and improving the recommending efficiency.

Description

Recommendation method, device, storage medium and equipment for user

Technical Field

The present invention relates to the field of data processing technologies, and in particular, to a recommendation method for a user, a recommendation apparatus for a user, a computer-readable storage medium, and a computer device.

Background

With the development of network information technology, people enter the era of information society and network economy, and various network services represented by electronic commerce are changing the lives and the works of people deeply. With the proliferation of electronic commerce, the types and quantity of services and goods that can be enjoyed through the network are increasing, but the time for consumers to browse and screen various services and goods is also greatly increased.

For enterprises, the cost of searching information is reduced for consumers, and the recommendation of services or commodities which are interesting and satisfactory to consumers to the consumers becomes an important marketing means. According to the characteristics of users, services or commodities which may be interested by the users are found and become hot spots of applications, but personalized recommendation for each user is low in efficiency, and in the face of more and more information push, consumers are not willing to believe the services and commodities directly recommended by an e-commerce platform, and the problem of low conversion rate of the recommended services and commodities is more and more prominent.

Disclosure of Invention

The embodiment of the invention provides a recommendation method and device, a computer readable storage medium and computer equipment for a user, which can recommend services or commodities which the user really has a demand, and improve the recommendation efficiency.

In order to solve the above problems, the present invention discloses a recommendation method for a user, comprising:

establishing an incidence relation between every two users according to the characteristic information of the users;

establishing at least one user group in all users, wherein any two users in the user group are directly associated or indirectly associated through at least one user;

determining characteristic users in the user group;

and recommending the network object by the user based on the characteristics.

Optionally, the feature information includes at least one of an attribute feature and a behavior feature of the user, where the attribute feature includes: identification, name, contact data; the behavior characteristics comprise: target users of the access behavior, submitted attribute information of other users, identification of connected networks or area identification.

Optionally, the step of establishing at least one user group among all users includes:

establishing a user relationship graph by taking each user as a node and taking the association relationship of each user pair with the same characteristic information as an edge;

searching a continuous and uninterrupted connected graph formed by a plurality of nodes from the user relationship graph;

and taking a plurality of users corresponding to all nodes in the connected graph as a user group.

Optionally, the step of establishing a user relationship graph includes:

acquiring each user identification and at least one corresponding characteristic information as node data of each user;

acquiring user identification and common characteristic information corresponding to the user pairs as side data of each user pair;

and establishing a user relationship graph consisting of the node data and the edge data.

Optionally, the edge data of the user relationship graph is distributed on a graph data server cluster, any two graph data servers in the graph data server cluster are directly associated or indirectly associated through at least one graph data server, and the step of searching a continuous and uninterrupted connected graph formed by a plurality of nodes from the user relationship graph includes:

calling each graph data server to respectively search at least one continuous and uninterrupted sub-connected graph formed by a plurality of nodes according to the edge data stored on each graph data server;

invoking each graph data server to distribute the sub-connected graph to at least one other graph data server associated with each graph data server;

and calling each graph data server to iteratively merge the searched or received sub-connected graphs, and distributing the sub-connected graphs to at least one other graph data server associated with each graph data server until no sub-connected graphs which can be merged exist.

Optionally, the step of determining the characteristic users in the user group includes:

for each user of the user group, counting association data representing the association relationship between the user and other users;

and determining the users with the associated data meeting the preset requirements as the characteristic users in the user group.

Optionally, the association data includes the number of users having an association relationship with the user, and the step of determining the user whose association data meets the preset requirement as the feature user in the user group includes:

and searching the user with the most association relation with other users, and determining the user as the characteristic user in the user group.

Optionally, the association data includes network behavior data, the network behavior data is data representing a degree of association between the user and another user, and the step of determining the user whose association data meets a preset requirement as the feature user in the user group includes:

acquiring network behavior data of each user in each user group;

and searching the users of which the network behavior data meet the preset requirements, and determining the users as the characteristic users in the user group.

Optionally, the step of recommending the network object by the user based on the characteristics includes:

acquiring all users accessing the network object;

searching the user group to which each user accessing the network object belongs;

and recommending the network object to the characteristic users in each searched user group.

acquiring all users accessing the network object;

searching at least one characteristic user of a user group in all users accessing the network object;

and recommending the network object to the non-characteristic users in the user group to which the characteristic users belong.

Optionally, the network object includes a network service, a data object.

Accordingly, the present invention discloses a recommendation apparatus for a user, comprising:

the relationship establishing module is used for establishing an incidence relationship between every two users according to the characteristic information of the users;

the group establishing module is used for establishing at least one user group in all users, and any two users in the user group are directly associated or indirectly associated through at least one user;

the characteristic user determining module is used for determining characteristic users in the user group;

and the network object recommending module is used for recommending the network object based on the characteristic user.

Optionally, the group establishing module includes:

the relation graph establishing submodule is used for establishing a user relation graph by taking each user as a node and taking the association relation of each user pair with the same characteristic information as an edge;

the connected graph searching submodule is used for searching a connected graph formed by a plurality of continuous and uninterrupted nodes from the user relationship graph;

and the group node submodule is used for taking a plurality of users corresponding to all nodes in the connected graph as a user group.

Optionally, the relationship graph establishing sub-module includes:

a node data acquiring unit, configured to acquire each user identifier and corresponding at least one piece of feature information as node data of each user;

a side data acquiring unit, configured to acquire a user identifier and common feature information corresponding to the user pair as side data of each user pair;

and the relationship graph establishing unit is used for establishing a user relationship graph formed by the node data and the edge data.

Optionally, the edge data of the user relationship graph is distributed on a graph data server cluster, any two graph data servers in the graph data server cluster are directly associated or indirectly associated through at least one graph data server, and the connected graph searching sub-module includes:

the sub-connected graph searching unit is used for calling each graph data server to respectively search at least one sub-connected graph formed by a plurality of continuous and uninterrupted nodes according to the edge data stored on the graph data server;

the sub-connected graph distributing unit is used for calling each graph data server to distribute the sub-connected graph to at least one other graph data server associated with each graph data server;

and the merging and distributing iteration unit is used for calling each graph data server to iterate and execute merging of the searched or received sub-connected graphs, and distributing the sub-connected graphs to at least one other graph data server associated with each graph data server until no sub-connected graphs which can be merged exist.

Optionally, the feature user determination module includes:

the associated data statistics submodule is used for counting associated data representing the association relation between the user and other users aiming at each user of the user group;

and the characteristic user determining submodule is used for determining the user of which the associated data meets the preset requirement as the characteristic user in the user group.

Optionally, the association data includes the number of users having an association relationship with the user, and the characteristic user determination sub-module is specifically configured to search the user having the most association relationship with other users, and determine the user as the characteristic user in the user group.

Optionally, the association data includes network behavior data, where the network behavior data is data representing a degree of association between the user and another user, and the characteristic user determination sub-module includes:

a behavior data acquiring unit, configured to acquire network behavior data of each user in each user group;

and the second characteristic user determining unit is used for searching the user of which the network behavior data meet the preset requirement and determining the user as the characteristic user in the user group.

Optionally, the network object recommendation module includes:

the first user acquisition submodule is used for acquiring all users accessing the network object;

the user group searching submodule is used for searching the user group to which each user accessing the network object belongs;

and the first network object recommending submodule is used for recommending the network object to the characteristic users in each searched user group.

Optionally, the network object recommendation module includes:

the second user acquisition submodule is used for acquiring all users accessing the network object;

the characteristic user searching sub-module is used for searching at least one characteristic user of a user group in all users accessing the network object;

and the second network object recommending submodule is used for recommending the network object to the non-characteristic users in the user group to which the characteristic users belong.

Accordingly, the present invention discloses a computer readable storage medium having a computer program stored thereon, characterized in that the program realizes the following steps when executed by a processor

determining characteristic users in the user group;

and recommending the network object by the user based on the characteristics.

Optionally, the step of establishing a user relationship graph includes:

acquiring network behavior data of each user in each user group;

acquiring all users accessing the network object;

Optionally, the network object includes a network service, a data object.

Accordingly, the present invention discloses a computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the following steps when executing the program:

determining characteristic users in the user group;

and recommending the network object by the user based on the characteristics.

Optionally, the step of establishing a user relationship graph includes:

acquiring network behavior data of each user in each user group;

acquiring all users accessing the network object;

Optionally, the network object includes a network service, a data object.

In summary, according to the embodiments of the present invention, the association relationship between two users is established according to the feature information of the users, at least one user group is established among all the users, the feature users in the user group are determined, and the network object is recommended based on the feature users, so that the users can be divided into different groups according to the feature information of the users, and the preferences of the users having the association relationship to the network object have certain similarity, and the recommendation service is performed for the user group, so that the services or goods really having the needs of the users can be recommended, and the recommendation efficiency is improved.

Drawings

FIG. 1 is a flowchart illustrating steps of a recommendation method for a user according to a first embodiment of the present invention;

FIG. 2 is a flowchart illustrating steps of a recommendation method for a user according to a second embodiment of the present invention;

fig. 3 is a block diagram illustrating a recommendation apparatus for a user according to a third embodiment of the present invention.

Detailed Description

In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.

Example one

Referring to fig. 1, a flowchart illustrating steps of a recommendation method for a user according to a first embodiment of the present invention is shown, which may specifically include the following steps:

step 101, establishing an association relationship between every two users according to the characteristic information of the users.

The feature information refers to data information of various attributes, behaviors and other features of the user, such as the name, telephone, mailbox, identification number, receiving address of the user, consignee information of a shopping order, identification of a connected network, personal information of the user filled in when booking a movie ticket, and the like. The present invention specifically includes any applicable various feature information, which is not limited in this respect.

In the embodiment of the invention, the characteristic information of all users is compared and analyzed, and the incidence relation between every two users is established according to the characteristic information of the users. The association relationship between two users means that two users have some association, for example, if the phone number of the reaper filled by one user during shopping is the same as the phone number of the other user, the two users have an association relationship. Any applicable association relationship is specifically included, and the embodiment of the present invention is not limited thereto.

For example, according to the feature information of all users, two users having the same feature information are searched, and an association relationship between the two users is established, where the association relationship may specifically record the identifiers of the two users and the same feature information.

At least one user group is established among all users, step 102.

In the embodiment of the invention, any two users in the user group are directly associated or indirectly associated through at least one user. After the incidence relation between every two users is established for all the users, at least one user group is established in all the users based on the incidence relation between every two users. Specifically, all user groups may be found out according to the association relationship between users, or a user group whose number of users reaches a certain number may be found out, or an arbitrary number of user groups may be found out.

In the embodiment of the present invention, the user group established in all the users may include all the users having direct or indirect associations, or may include some users having direct or indirect associations, which is not limited in the embodiment of the present invention.

And 103, determining characteristic users in the user group.

In this embodiment of the present invention, the feature users refer to a part of users selected from a user group, and specifically, the feature users may be determined according to the number of association relationships established with other users, or the feature users may be determined according to the number of comments or browsing numbers of the users, or the feature users may be determined according to basic information of the users, or the feature users may be determined according to any applicable manner, which is not limited in this embodiment of the present invention.

For example, the user subjected to authentication is determined as the characteristic user, or the user with the highest incidence relation established with other users in each user group is determined as the characteristic user, or the user with the highest comment or browsing number of the text message is determined as the characteristic user.

And 104, recommending the network object by the user based on the characteristics.

In the embodiment of the present invention, the network object refers to any service or object that can be accessed through a network, the service may be at least one of an air ticket booking, a hotel booking, a take-out and the like, and the object may be at least one of a commodity, a picture, a video, music, a novel and the like.

In the embodiment of the present invention, the recommendation of the network object based on the feature user may specifically be to recommend the network object to the feature user in the user group, or recommend the network object accessed by the feature user to another user of the feature-removed user in the user group, and specifically, the recommendation may be performed based on the feature user in any applicable manner, which is not limited in this embodiment of the present invention.

In summary, according to the embodiments of the present invention, the association relationship between two users is established according to the feature information of the users, at least one user group is established among all the users, the feature users in the user group are determined, and the recommendation of the network object is performed based on the feature users, so that the users can be divided into different groups according to the feature information of the users, and the preferences of the users having the association relationship to the network object have a certain similarity, so that the recommendation service is performed for the user group, and the services or goods really having the needs of the users can be recommended, thereby improving the recommendation efficiency.

In a preferred embodiment of the present invention, the feature information includes at least one of attribute features and behavior features of the user, where the attribute features include: identification, name, contact data; the behavior characteristics comprise: target users of the access behavior, submitted attribute information of other users, identification of connected networks or area identification.

The attribute features refer to features related to user attributes, and may specifically be features closely related to user identities, including identifiers, names, contact data, and the like. The behavior characteristics refer to characteristics related to the behavior of the user in the network, and include target users accessing the behavior, attribute information of other submitted users, and identification of connected networks or area identification.

The access behavior refers to the behaviors of contact, click, download, purchase, trial and the like on the network platform, and target users of the access behavior can include contacted parties, users corresponding to the clicked and accessed head portrait or personal page, users of downloaded and uploaded files, purchased parties, tried parties and the like. The submitted attribute information of other users refers to the attribute information such as identification, name, contact information and the like of other users filled when the users submit any information on the network platform. The identifier or area identifier of the connected network refers to a name identifier or area name of the connected wireless local area network, an address or network segment of the connected wired network, and the like.

According to the various characteristic information, the association relationship among various users can be found, and the association relationship among the users can be established as long as the users are in contact with each other.

In a preferred embodiment of the present invention, the step of determining the characteristic users in the user group comprises: and for each user of the user group, counting the associated data representing the association relationship between the user and other users, and determining the user with the associated data meeting the preset requirements as the characteristic user in the user group.

The association data is data that can represent an association relationship between users, for example, the number of users who establish an association relationship with a user, the number of text messages posted by a user and reviewed by other users, and the like, and may specifically include any applicable association data, which is not limited in this embodiment of the present invention. And for each user group, counting the associated data representing the association relationship between each user in the user group and other users in the group.

And judging whether the associated data of each user meets the preset requirement or not aiming at each user group, and determining the user of which the associated data meets the preset requirement as a characteristic user in the user group. For example, the user with the most association relationship with other users in each user group is determined as the characteristic user, or the user with the most number of comments or browsed text messages is determined as the characteristic user.

The users with the associated data meeting the preset requirements are determined as the characteristic users in the user group, the characteristic users with more associated with other users can be found, the characteristic users can recommend the network object to more users in the user group and further spread the network object to more users, the recommendation efficiency can be improved by recommending the characteristic users, the problem that the consumers are unwilling to trust the service and the commodity directly recommended by the platform is solved, the network object is recommended to the characteristic users, the characteristic users can recommend the network object again in the user group, and the other users are more willing to accept the recommended content, so that the recommendation conversion rate is improved.

In a preferred embodiment of the present invention, the association data includes network behavior data, the network behavior data is data representing a degree of association between the user and another user, and the step of determining a user whose association data meets a preset requirement as a feature user in the user group includes: acquiring network behavior data of each user in each user group; and searching the users of which the network behavior data meet the preset requirements, and determining the users as the characteristic users in the user group.

The associated data may include network behavior data, the network behavior data is data representing the degree of association between the user and other users, for example, the number of comments made on the teletext messages posted by the user, the number of browsed teletext messages, and the like, and the network behavior data may be a statistical value, for example, the sum of the number of comments made on all teletext messages posted by the user.

For each user group, network behavior data of each user is obtained, and users whose network behavior data meet preset requirements are searched, for example, the number of the users whose comments in the user group exceeds ten thousand is a preset requirement, and thus the searched users are all users whose number of the users whose comments in the user group exceeds ten thousand. Any suitable preset requirement is adopted according to actual requirements, and the embodiment of the invention is not limited to this. In this way, the users with the most influential or transmissible power among the user population are found. And determining the searched users as the characteristic users in the user group.

In a preferred embodiment of the present invention, the step of recommending the network object by the user based on the feature comprises: acquiring all users accessing the network object; searching at least one characteristic user of a user group in all users accessing the network object; and recommending the network object to the non-characteristic users in the user group to which the characteristic users belong.

Accessing the network object refers to acquiring all users accessing the network object from historical access data of the network object aiming at various network behaviors of purchasing, reserving, clicking, downloading, using and the like of the network object, and specifically may be acquiring identifications of all users. The characteristic users of at least one user group are searched among all users accessing the network object. When the network object is recommended, users which are characteristic users in all users of the network object are found, and the network object is recommended to non-characteristic users in a user group to which each characteristic user belongs. Because the preferences of the users belonging to the same user group to the network object have similarity, and the characteristic users have relevance with more users in the user group, the similarity between the network object accessed by the characteristic users and the preferences of more users in the user group is stronger, the network object accessed by the characteristic users is recommended to the non-characteristic users in the user group, and the recommendation conversion rate can be improved.

In a preferred embodiment of the invention, the network object comprises a network service, a data object. The network service refers to a service provided through a network, such as a service of booking an air ticket, booking a hotel, booking a take-out and the like. The data object refers to an object such as a commodity, a video, a document, etc., provided through a network, for example, a toy, an episode, a novel, etc., which is sold.

Referring to fig. 2, a flowchart illustrating steps of a recommendation method for a user according to a second embodiment of the present invention is shown, which may specifically include the following steps:

step 201, establishing an association relationship between every two users according to the characteristic information of the users.

In the embodiment of the present invention, reference may be made to the description of the foregoing embodiment for specific implementation of this step, which is not described herein again.

Step 202, establishing a user relationship graph by taking each user as a node and taking the association relationship of each user pair with the same characteristic information as an edge.

In the embodiment of the invention, the user relationship graph refers to a relationship graph which comprises all users and connects all users by the incidence relation among the users. The user relationship graph stores nodes and edges, and specifically, each user is taken as a node, and the association relationship of each user pair with the same characteristic information is taken as an edge. The nodes in the user relationship graph may store data including user identifiers and feature information, and the edges in the user relationship graph may store the identifiers of users at both ends of one edge and common feature information.

In a preferred embodiment of the present invention, the step of establishing the user relationship graph comprises: acquiring each user identification and at least one corresponding characteristic information as node data of each user; acquiring user identification and common characteristic information corresponding to the user pairs as side data of each user pair; and establishing a user relationship graph consisting of the node data and the edge data.

The node data refers to data comprising user identifications and corresponding at least one piece of characteristic information, the side data refers to data comprising user identifications corresponding to the users and common characteristic information, and then the node data and the side data form a user relation graph.

Specifically, a user relationship graph may be established on the graph data server cluster, and all node data are first uniformly split and respectively sent to each analysis server of the graph data server cluster for storage. And searching node pairs with the same characteristic information according to the node data, and generating the edge data of each user pair according to the user identification and the common characteristic information corresponding to the two nodes of the node pairs. In specific implementation, each graph data server may be called to respectively search for a target node having the same characteristic information as each node included in the node data stored on the graph data server, and the node data of the target node may be stored on any graph data server. And calling each graph data server to acquire the user identification and the common characteristic information corresponding to the two nodes of the node pair searched by the graph data server. And calling each graph data server to store the user identification and the common characteristic information of the acquired node pairs as the edge data of each user pair.

After the edge data of each user pair is generated according to the user identifiers and the common characteristic information corresponding to the two nodes of the node pair, all the edge data with the same user identifiers and the common characteristic information can be searched in the graph data server cluster, and repeated edge data are deleted until the edge data stored in the graph data server cluster are not repeated. And uniformly segmenting the edge data stored on each graph data server, and distributing the segmented edge data to each graph data server in the graph data server cluster, thereby completing the establishment of the user relationship graph in the graph data server cluster.

Step 203, searching a continuous and uninterrupted connected graph formed by a plurality of nodes from the user relationship graph.

In the embodiment of the present invention, a connected graph formed by a plurality of continuous and uninterrupted nodes is searched from the user relationship graph, and any two nodes in the connected graph may be connected directly or through at least one other node. The specifically searched connectivity graph may be a maximum connectivity graph or may not be a maximum connectivity graph, which is not limited in this embodiment of the present invention.

In a preferred embodiment of the present invention, the edge data of the user relationship graph is distributed on a graph data server cluster, any two graph data servers in the graph data server cluster are directly associated or indirectly associated through at least one graph data server, and the step of searching a connected graph formed by a plurality of continuous and uninterrupted nodes from the user relationship graph may include: calling each graph data server to respectively search at least one continuous and uninterrupted sub-connected graph formed by a plurality of nodes according to the edge data stored on each graph data server; invoking each graph data server to distribute the sub-connected graph to at least one other graph data server associated with each graph data server; and calling each graph data server to iteratively merge the searched or received sub-connected graphs, and distributing the sub-connected graphs to at least one other graph data server associated with each graph data server until no sub-connected graphs which can be merged exist.

Because the number of users is huge, the data volume of the user relation graph is too large, the graph data cannot meet the requirement when a single server is used for processing the graph data, and the graph data related operation can be carried out on a graph data server cluster. Specifically, the edge data of the user relationship graph is distributed on a graph data server cluster, and any two graph data servers in the graph data server cluster are directly associated or indirectly associated through at least one graph data server.

In specific implementation, each graph data server is called, and at least one sub-connected graph formed by a plurality of continuous and uninterrupted nodes is searched according to the edge data stored on the graph data server, wherein each sub-connected graph is obtained only by each graph data server according to the edge data stored by the sub-connected graph, and communication with other graph data servers is not generated. And then, calling each graph data server to distribute the sub-connected graph to at least one other graph data server associated with each graph data server, specifically to distribute the sub-connected graph and the edge data in the sub-connected graph to other graph data servers. After receiving the sub-connected graphs sent by other servers, each graph data server merges the sub-connected graphs with the same nodes into one according to the searched sub-connected graph and the received sub-connected graph and merging, then distributes the sub-connected graphs again, and continuously and iteratively executes merging of the searched or received sub-connected graphs and distributes the sub-connected graphs to at least one other graph data server associated with each graph data server until no sub-connected graphs which can be merged exist. Because any two graph data servers in the graph data server cluster are directly associated or indirectly associated through at least one graph data server, after multiple iterations, the graph data server cluster obtains a connected graph in the user relationship graph. The work of searching the connected graph can be carried out on different machines in parallel, the result on each machine is only needed to be gathered, the network overhead is low, and the operation efficiency is improved.

And step 204, taking a plurality of users corresponding to all nodes in the connected graph as a user group.

In the embodiment of the invention, after the connected graph is obtained, each node in the connected graph is obtained, and a plurality of users corresponding to all the nodes in the connected graph are used as a user group.

Step 205, for each user of the user group, statistics is performed on the associated data representing the association relationship between the user and other users.

And step 206, searching the user with the most association relation with other users, and determining the user as the characteristic user in the user group.

In the embodiment of the present invention, the association data includes the number of users having an association relationship with the user. And aiming at each user group, searching the users with the most incidence relation with other users, and determining the searched users as characteristic users.

Step 207, obtain all users accessing the network object.

In the embodiment of the invention, for the network object to be recommended, all users who have access to the network object, for example, all users who have used the airline ticket booking service in the platform, are obtained.

Step 208, the user group to which each user accessing the network object belongs is searched.

In the embodiment of the invention, the user group to which each user accessing the network object belongs is searched according to the searched user groups.

And step 209, recommending the network object to the characteristic users in each searched user group.

In the embodiment of the invention, the network object is recommended to the characteristic users in each searched user group, for example, the service of booking air tickets is recommended to the characteristic users in the user group.

To sum up, according to the embodiment of the present invention, the association relationship between two users is established according to the feature information of the users, each user is taken as a node, the association relationship of each user pair with the same feature information is taken as an edge, a user relationship graph is established, a continuous connection graph formed by a plurality of nodes is searched from the user relationship graph, a plurality of users corresponding to all nodes in the connection graph are taken as a user group, so that the relationship graph of the users can be established according to the feature information of the users to find each association relationship between the users, then the users are divided into different groups by searching the connection graph based on the relationship graph of the users, the preferences of the users with the association relationship to the network object have certain similarity, the users are recommended for the user group, and the services or goods really needed by the users can be recommended, the efficiency of recommendation is improved.

Further, by counting the associated data representing the association relationship between the user and other users for each user of the user group, searching the user with the most association relationship with other users, determining the user as a characteristic user in the user group, acquiring all users accessing a network object, searching the user group to which each user accessing the network object belongs, recommending the network object to the characteristic user in each searched user group, so that the characteristic users with more association with other users can be searched, the characteristic users can recommend the network object to more users in the user group and further spread the network object to more users, the efficiency of recommendation can be improved by recommending the characteristic users, the problem that consumers are not willing to trust the service and goods directly recommended by the platform is solved, and the network object is recommended to the characteristic users, and then the characteristic users recommend again in the user group, and other users are more willing to accept the recommended content, thereby improving the conversion rate of recommendation.

It should be noted that, for simplicity of description, the method embodiments are described as a series of acts or combination of acts, but those skilled in the art will recognize that the present invention is not limited by the illustrated order of acts, as some steps may occur in other orders or concurrently in accordance with the embodiments of the present invention. Further, those skilled in the art will appreciate that the embodiments described in the specification are presently preferred and that no particular act is required to implement the invention.

EXAMPLE III

Referring to fig. 3, a block diagram of a structure of a recommendation device for a user according to a third embodiment of the present invention is shown, and specifically includes the following modules:

the relationship establishing module 301 is configured to establish an association relationship between every two users according to the feature information of the users;

a group establishing module 302, configured to establish at least one user group among all users, where any two users in the user group are directly associated or indirectly associated through at least one user;

a characteristic user determining module 303, configured to determine a characteristic user in the user group;

and a network object recommending module 304, configured to recommend a network object based on the feature user.

In the embodiment of the present invention, preferably, the feature information includes at least one of an attribute feature and a behavior feature of the user, where the attribute feature includes: identification, name, contact data; the behavior characteristics comprise: target users of the access behavior, submitted attribute information of other users, identification of connected networks or area identification.

In the embodiment of the present invention, preferably, the group establishing module includes:

In the embodiment of the present invention, preferably, the relationship diagram establishing sub-module includes:

In this embodiment of the present invention, preferably, the edge data of the user relationship graph is distributed on a graph data server cluster, any two graph data servers in the graph data server cluster are directly associated or indirectly associated through at least one graph data server, and the connected graph search sub-module includes:

In the embodiment of the present invention, preferably, the characteristic user determining module includes:

In the embodiment of the present invention, preferably, the association data includes the number of users having an association relationship with a user, and the characteristic user determination sub-module is specifically configured to search a user having the most association relationship with other users, and determine the user as a characteristic user in the user group.

In this embodiment of the present invention, preferably, the association data includes network behavior data, where the network behavior data is data representing a degree of association between the user and another user, and the characteristic user determination sub-module includes:

In the embodiment of the present invention, preferably, the network object recommending module includes:

In the embodiment of the present invention, preferably, the network object includes a network service and a data object.

Example four

The present embodiment provides a computer-readable storage medium having a computer program stored thereon, wherein the program, when executed by a processor, performs the steps of:

determining characteristic users in the user group;

and recommending the network object by the user based on the characteristics.

In the embodiment of the present invention, preferably, the step of establishing at least one user group among all users includes:

In the embodiment of the present invention, preferably, the step of establishing the user relationship graph includes:

In this embodiment of the present invention, preferably, the edge data of the user relationship graph is distributed on a graph data server cluster, any two graph data servers in the graph data server cluster are directly associated or indirectly associated through at least one graph data server, and the step of searching a connected graph formed by a plurality of continuous and uninterrupted nodes from the user relationship graph includes:

In this embodiment of the present invention, preferably, the step of determining the characteristic users in the user group includes:

In the embodiment of the present invention, preferably, the association data includes the number of users having an association relationship with the user, and the step of determining the user whose association data meets a preset requirement as the feature user in the user group includes:

In the embodiment of the present invention, preferably, the association data includes network behavior data, the network behavior data is data representing a degree of association between a user and another user, and the step of determining a user whose association data meets a preset requirement as a feature user in the user group includes:

acquiring network behavior data of each user in each user group;

In this embodiment of the present invention, preferably, the step of recommending the network object by the user based on the characteristics includes:

acquiring all users accessing the network object;

EXAMPLE five

The embodiment provides a computer device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the following steps:

determining characteristic users in the user group;

and recommending the network object by the user based on the characteristics.

acquiring network behavior data of each user in each user group;

acquiring all users accessing the network object;

For the device embodiment, since it is basically similar to the method embodiment, the description is simple, and for the relevant points, refer to the partial description of the method embodiment.

The embodiments in the present specification are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, apparatus, or computer program product. Accordingly, embodiments of the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, embodiments of the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing terminal to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing terminal to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing terminal to cause a series of operational steps to be performed on the computer or other programmable terminal to produce a computer implemented process such that the instructions which execute on the computer or other programmable terminal provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

While preferred embodiments of the present invention have been described, additional variations and modifications of these embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the embodiments of the invention.

Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other like elements in a process, method, article, or terminal that comprises the element.

The recommendation method, apparatus, device and storage medium for users provided by the present invention are described in detail above, and the principle and implementation of the present invention are explained in this document by applying specific examples, and the description of the above embodiments is only used to help understanding the method and core ideas of the present invention; meanwhile, for a person 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. A recommendation method for a user, comprising:

determining characteristic users in the user group;

recommending network objects based on the characteristic users;

wherein the feature users comprise part of users in the user group, and the network object comprises a service or an object accessed through a network;

the recommending of the network object based on the characteristic user comprises the following steps:

recommending the network object to the feature user; or

Recommending the network object accessed by the characteristic user to other users of the characteristic-removed user in the user group;

the characteristic information comprises at least one of attribute characteristics and behavior characteristics of the user, wherein the attribute characteristics comprise: identification, name, contact data; the behavior characteristics comprise: target users of the access behavior, attribute information of other submitted users, and identifiers or area identifiers of connected networks;

the step of determining the characteristic users in the user group comprises:

2. The method of claim 1, wherein the step of establishing at least one user group among all users comprises:

3. The method of claim 2, wherein the step of establishing a user relationship graph comprises:

4. The method according to claim 3, wherein the edge data of the user relationship graph is distributed on a graph data server cluster, any two graph data servers in the graph data server cluster are directly associated or indirectly associated through at least one graph data server, and the step of finding a connected graph formed by a plurality of continuous and uninterrupted nodes from the user relationship graph comprises:

5. The method according to claim 1, wherein the association data includes the number of users having an association relationship with the user, and the step of determining the user whose association data meets a preset requirement as the characteristic user in the user group includes:

6. The method according to claim 1, wherein the association data includes network behavior data, the network behavior data is data representing the degree of association between the user and other users, and the step of determining the user whose association data meets preset requirements as the characteristic user in the user group includes:

acquiring network behavior data of each user in each user group;

7. The method of claim 1, wherein the step of recommending network objects based on the feature users comprises:

acquiring all users accessing the network object;

8. The method of claim 1, wherein the step of recommending network objects based on the feature users comprises:

acquiring all users accessing the network object;

9. The method of claim 1, wherein the network object comprises a web service, a data object.

10. A recommendation device for a user, comprising:

the network object recommending module is used for recommending network objects based on the characteristic users;

the network object recommendation module is further configured to:

recommending the network object to the feature user; or

the feature user determination module includes:

11. A computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, carries out the method according to any one of claims 1 to 9.

12. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor when executing the program implements the method of any one of claims 1 to 9.