CN108416645A - A kind of recommendation method, apparatus, storage medium and equipment for user - Google Patents
A kind of recommendation method, apparatus, storage medium and equipment for user Download PDFInfo
- Publication number
- CN108416645A CN108416645A CN201810054377.3A CN201810054377A CN108416645A CN 108416645 A CN108416645 A CN 108416645A CN 201810054377 A CN201810054377 A CN 201810054377A CN 108416645 A CN108416645 A CN 108416645A
- Authority
- CN
- China
- Prior art keywords
- user
- feature
- users
- group
- user group
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
Abstract
An embodiment of the present invention provides a kind of recommendation method, apparatus, storage medium and computer equipments for user.The method includes the incidence relation between two two users is established according to the characteristic information of user, at least one user group is established in all users, determine the feature user in the user group, the recommendation of network object is carried out based on the feature user, allow to the characteristic information according to user, user is divided into different groups, and the user with incidence relation to the preference of network object with certain similitude, recommendation service is carried out for user group, service or commodity of the user really with demand can be more recommended to, the efficiency of recommendation is improved.
Description
Technical field
The present invention relates to technical field of data processing, are directed to more particularly to a kind of recommendation method, one kind for user
The recommendation apparatus of user, a kind of computer readable storage medium and a kind of computer equipment.
Background technology
With the development of the network information technology, people enter information-intensive society and the age of Internet economy, are with e-commerce
The multiple network service of representative is just changing people’s lives and work deeply.With the prosperity of e-commerce, pass through network institute
The type and quantity of the service and commodity that can enjoy are more and more, but also considerably increase consumer browsing, screen it is a variety of
The time of various service and commodity.
For enterprise, the cost for searching information, service consumer is interested, satisfied or quotient are reduced for consumer
Product, which recommend consumer, becomes highly important marketing methods.According to the feature of user, it is found that user may interested clothes
Business or commodity become the hot spot of application, but the recommendation efficiency that personalization is carried out for each user is very low, and in face of more
Carry out more information push, consumer becomes to be unwilling to believe the service and commodity that e-commerce platform is directly recommended, recommendation
The service problem low with commodity conversion rate is more and more prominent.
Invention content
The embodiment of the present invention provides a kind of recommendation method, apparatus, computer readable storage medium and calculating for user
Machine equipment can more recommend to service or commodity of the user really with demand, improve the efficiency of recommendation.
To solve the above-mentioned problems, the invention discloses a kind of recommendation methods for user, including:
The incidence relation between two two users is established according to the characteristic information of user;
Establish at least one user group in all users, in the user group any two user be directly linked or
Pass through at least one user's indirect association;
Determine the feature user in the user group;
The recommendation of network object is carried out based on the feature user.
Optionally, the characteristic information includes at least one of attributive character and behavioural characteristic of user, wherein attribute is special
Sign includes:Mark, title, contact method data;Behavioural characteristic includes:The target user of access behavior, the other users of submission
Attribute information, the network connected mark or area identification.
Optionally, described the step of establishing at least one user group in all users, includes:
Using each user as node, using the incidence relation of each user couple with same characteristic features information as side, establish use
Family relational graph;
The connected graph that continuous continual multiple nodes are constituted is searched from the customer relationship figure;
The corresponding multiple users of all nodes are as a user group using in connected graph.
Optionally, described the step of establishing customer relationship figure, includes:
Obtain the node data of each user identifier and corresponding at least one characteristic information as each user;
Obtain number of edges evidence of the user to corresponding user identifier and common trait information as each user couple;
It establishes by the node data and number of edges according to the customer relationship figure formed.
Optionally, the number of edges evidence of the customer relationship figure is distributed on diagram data server cluster, the diagram data service
In device cluster any two diagram data server be directly linked or by least one diagram data server indirect association, it is described
The step of connected graph of the continuous continual multiple nodes compositions of lookup, includes from the customer relationship figure:
It calls each diagram data server according to the number of edges evidence stored thereon, searches continuous continual multiple sections respectively
At least one sub- connected graph that point is constituted;
Each diagram data server is called to be distributed to the sub- connected graph associated at least one with each diagram data server
Other diagram data servers;
Each diagram data server iteration is called to execute the sub- connected graph for merging and finding or receiving, and even by the son
Logical figure be distributed to the associated at least one other diagram data server of each diagram data server, until it is annexable son even
Logical figure.
Optionally, the step of feature user in the determination user group includes:
For each user of the user group, the statistics characterization user is associated with the incidence relation of other users
Data;
The user that associated data is met to preset requirement is determined as feature user in the user group.
Optionally, the associated data includes the quantity for the user for having incidence relation with user, described by associated data
The step of meeting the feature user that the user of preset requirement is determined as in the user group include:
Search the user for having incidence relation most with other users, the feature user being determined as in the user group.
Optionally, the associated data includes network behavior data, and the network behavior data are characterization user and other
The data of user-association degree, the user that associated data is met to preset requirement are determined as the feature in the user group
The step of user includes:
In each user group, the network behavior data of each user are obtained;
The user that the network behavior data reach preset requirement is searched, the feature being determined as in the user group is used
Family.
Optionally, it is described based on the feature user carry out network object recommendation the step of include:
Obtain all users for accessing network object;
Search the user group for each user attaching for accessing network object;
The network object is recommended to the feature user in the user group respectively found.
Optionally, it is described based on the feature user carry out network object recommendation the step of include:
Obtain all users for accessing network object;
The feature user of at least one user group is searched in all users for accessing network object;
The network object is recommended to the non-feature user in the user group of each feature user attaching.
Optionally, the network object includes network service, data object.
Correspondingly, the invention discloses a kind of recommendation apparatus for user, including:
Relationship establishes module, for establishing the incidence relation between two two users according to the characteristic information of user;
Group establishes module, arbitrary in the user group for establishing at least one user group in all users
Two users are directly linked or pass through at least one user's indirect association;
Feature user's determining module, for determining the feature user in the user group;
Network object recommending module, the recommendation for carrying out network object based on the feature user.
Optionally, the characteristic information includes at least one of attributive character and behavioural characteristic of user, wherein attribute is special
Sign includes:Mark, title, contact method data;Behavioural characteristic includes:The target user of access behavior, the other users of submission
Attribute information, the network connected mark or area identification.
Optionally, the group establishes module and includes:
Relational graph setting up submodule is used for using each user as node, with each user couple with same characteristic features information
Incidence relation be side, establish customer relationship figure;
Connected graph searches submodule, for searching continuous continual multiple nodes compositions from the customer relationship figure
Connected graph;
Group's node submodule, for using in connected graph the corresponding multiple users of all nodes as a user group.
Optionally, the relational graph setting up submodule includes:
Node data acquiring unit, for obtaining each user identifier and corresponding at least one characteristic information as each
The node data of a user;
Side data capture unit, for obtaining the user to corresponding user identifier and common trait information as each
The number of edges evidence of user couple;
Relational graph establishes unit, for establishing by the node data and number of edges according to the customer relationship figure formed.
Optionally, the number of edges evidence of the customer relationship figure is distributed on diagram data server cluster, the diagram data service
In device cluster any two diagram data server be directly linked or by least one diagram data server indirect association, it is described
Connected graph searches submodule:
Sub- connected graph searching unit, for calling each diagram data server according to the number of edges evidence stored thereon, respectively
Search at least one sub- connected graph that continuous continual multiple nodes are constituted;
Sub- connected graph Dispatching Unit, for calling each diagram data server to be distributed to the sub- connected graph and each diagram data
The associated at least one other diagram data server of server;
Merge and distribute iteration unit, for calling each diagram data server iteration to execute what merging found or received
Sub- connected graph, and the sub- connected graph is distributed to and the associated at least one other diagram data service of each diagram data server
Device, until without annexable sub- connected graph.
Optionally, the feature user determining module includes:
Associated data statistic submodule, for each user for the user group, statistics characterize the user with
The associated data of the incidence relation of other users;
Feature user's determination sub-module, the user for associated data to be met to preset requirement are determined as the user group
In feature user.
Optionally, the associated data includes the quantity for the user for having incidence relation with user, and the feature user is true
Stator modules are specifically used for searching the user for having incidence relation most with other users, be determined as in the user group
Feature user.
Optionally, the associated data includes network behavior data, and the network behavior data are characterization user and other
The data of user-association degree, the feature user determination sub-module include:
Behavioral data acquiring unit, in each user group, obtaining the network behavior data of each user;
Second feature user's determination unit reaches the user of preset requirement for searching the network behavior data, determines
For the feature user in the user group.
Optionally, the network object recommending module includes:
First user's acquisition submodule, for obtaining all users for accessing network object;
User group searches submodule, the user group for searching each user attaching for accessing network object;
First network object recommendation submodule, for recommending the network object in the user group respectively found
Feature user.
Optionally, the network object recommending module includes:
Second user acquisition submodule, for obtaining all users for accessing network object;
Feature user searches submodule, for searching at least one user group in all users for accessing network object
Feature user;
Second network object recommends submodule, the user group for the network object to be recommended to each feature user attaching
Non- feature user in body.
Correspondingly, the invention discloses a kind of computer readable storage mediums, are stored thereon with computer program, feature
It is, which realizes following steps when being executed by processor
The incidence relation between two two users is established according to the characteristic information of user;
Establish at least one user group in all users, in the user group any two user be directly linked or
Pass through at least one user's indirect association;
Determine the feature user in the user group;
The recommendation of network object is carried out based on the feature user.
Optionally, the characteristic information includes at least one of attributive character and behavioural characteristic of user, wherein attribute is special
Sign includes:Mark, title, contact method data;Behavioural characteristic includes:The target user of access behavior, the other users of submission
Attribute information, the network connected mark or area identification.
Optionally, described the step of establishing at least one user group in all users, includes:
Using each user as node, using the incidence relation of each user couple with same characteristic features information as side, establish use
Family relational graph;
The connected graph that continuous continual multiple nodes are constituted is searched from the customer relationship figure;
The corresponding multiple users of all nodes are as a user group using in connected graph.
Optionally, described the step of establishing customer relationship figure, includes:
Obtain the node data of each user identifier and corresponding at least one characteristic information as each user;
Obtain number of edges evidence of the user to corresponding user identifier and common trait information as each user couple;
It establishes by the node data and number of edges according to the customer relationship figure formed.
Optionally, the number of edges evidence of the customer relationship figure is distributed on diagram data server cluster, the diagram data service
In device cluster any two diagram data server be directly linked or by least one diagram data server indirect association, it is described
The step of connected graph of the continuous continual multiple nodes compositions of lookup, includes from the customer relationship figure:
It calls each diagram data server according to the number of edges evidence stored thereon, searches continuous continual multiple sections respectively
At least one sub- connected graph that point is constituted;
Each diagram data server is called to be distributed to the sub- connected graph associated at least one with each diagram data server
Other diagram data servers;
Each diagram data server iteration is called to execute the sub- connected graph for merging and finding or receiving, and even by the son
Logical figure be distributed to the associated at least one other diagram data server of each diagram data server, until it is annexable son even
Logical figure.
Optionally, the step of feature user in the determination user group includes:
For each user of the user group, the statistics characterization user is associated with the incidence relation of other users
Data;
The user that associated data is met to preset requirement is determined as feature user in the user group.
Optionally, the associated data includes the quantity for the user for having incidence relation with user, described by associated data
The step of meeting the feature user that the user of preset requirement is determined as in the user group include:
Search the user for having incidence relation most with other users, the feature user being determined as in the user group.
Optionally, the associated data includes network behavior data, and the network behavior data are characterization user and other
The data of user-association degree, the user that associated data is met to preset requirement are determined as the feature in the user group
The step of user includes:
In each user group, the network behavior data of each user are obtained;
The user that the network behavior data reach preset requirement is searched, the feature being determined as in the user group is used
Family.
Optionally, it is described based on the feature user carry out network object recommendation the step of include:
Obtain all users for accessing network object;
Search the user group for each user attaching for accessing network object;
The network object is recommended to the feature user in the user group respectively found.
Optionally, it is described based on the feature user carry out network object recommendation the step of include:
Obtain all users for accessing network object;
The feature user of at least one user group is searched in all users for accessing network object;
The network object is recommended to the non-feature user in the user group of each feature user attaching.
Optionally, the network object includes network service, data object.
Correspondingly, the invention discloses a kind of computer equipment, including memory, processor and storage are on a memory simultaneously
The computer program that can be run on a processor, which is characterized in that the processor realizes following steps when executing described program:
The incidence relation between two two users is established according to the characteristic information of user;
Establish at least one user group in all users, in the user group any two user be directly linked or
Pass through at least one user's indirect association;
Determine the feature user in the user group;
The recommendation of network object is carried out based on the feature user.
Optionally, the characteristic information includes at least one of attributive character and behavioural characteristic of user, wherein attribute is special
Sign includes:Mark, title, contact method data;Behavioural characteristic includes:The target user of access behavior, the other users of submission
Attribute information, the network connected mark or area identification.
Optionally, described the step of establishing at least one user group in all users, includes:
Using each user as node, using the incidence relation of each user couple with same characteristic features information as side, establish use
Family relational graph;
The connected graph that continuous continual multiple nodes are constituted is searched from the customer relationship figure;
The corresponding multiple users of all nodes are as a user group using in connected graph.
Optionally, described the step of establishing customer relationship figure, includes:
Obtain the node data of each user identifier and corresponding at least one characteristic information as each user;
Obtain number of edges evidence of the user to corresponding user identifier and common trait information as each user couple;
It establishes by the node data and number of edges according to the customer relationship figure formed.
Optionally, the number of edges evidence of the customer relationship figure is distributed on diagram data server cluster, the diagram data service
In device cluster any two diagram data server be directly linked or by least one diagram data server indirect association, it is described
The step of connected graph of the continuous continual multiple nodes compositions of lookup, includes from the customer relationship figure:
It calls each diagram data server according to the number of edges evidence stored thereon, searches continuous continual multiple sections respectively
At least one sub- connected graph that point is constituted;
Each diagram data server is called to be distributed to the sub- connected graph associated at least one with each diagram data server
Other diagram data servers;
Each diagram data server iteration is called to execute the sub- connected graph for merging and finding or receiving, and even by the son
Logical figure be distributed to the associated at least one other diagram data server of each diagram data server, until it is annexable son even
Logical figure.
Optionally, the step of feature user in the determination user group includes:
For each user of the user group, the statistics characterization user is associated with the incidence relation of other users
Data;
The user that associated data is met to preset requirement is determined as feature user in the user group.
Optionally, the associated data includes the quantity for the user for having incidence relation with user, described by associated data
The step of meeting the feature user that the user of preset requirement is determined as in the user group include:
Search the user for having incidence relation most with other users, the feature user being determined as in the user group.
Optionally, the associated data includes network behavior data, and the network behavior data are characterization user and other
The data of user-association degree, the user that associated data is met to preset requirement are determined as the feature in the user group
The step of user includes:
In each user group, the network behavior data of each user are obtained;
The user that the network behavior data reach preset requirement is searched, the feature being determined as in the user group is used
Family.
Optionally, it is described based on the feature user carry out network object recommendation the step of include:
Obtain all users for accessing network object;
Search the user group for each user attaching for accessing network object;
The network object is recommended to the feature user in the user group respectively found.
Optionally, it is described based on the feature user carry out network object recommendation the step of include:
Obtain all users for accessing network object;
The feature user of at least one user group is searched in all users for accessing network object;
The network object is recommended to the non-feature user in the user group of each feature user attaching.
Optionally, the network object includes network service, data object.
In conclusion embodiment according to the present invention, by establishing the pass between two two users according to the characteristic information of user
Connection relationship establishes at least one user group in all users, determines the feature user in the user group, based on described
Feature user carries out the recommendation of network object so that user can be divided into different groups according to the characteristic information of user,
And the user with incidence relation to the preference of network object with certain similitude, carry out recommendation clothes for user group
Business can more recommend to service or commodity of the user really with demand, improve the efficiency of recommendation.
Description of the drawings
Fig. 1 shows the step flow chart of a kind of according to embodiments of the present invention one recommendation method for user;
Fig. 2 shows the step flow charts of a kind of according to embodiments of the present invention two recommendation method for user;
Fig. 3 shows a kind of structure diagram of according to embodiments of the present invention three recommendation apparatus for user.
Specific implementation mode
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, below in conjunction with the accompanying drawings and specific real
Applying mode, the present invention is described in further detail.
Embodiment one
Referring to Fig.1, the step flow chart of a kind of according to embodiments of the present invention one recommendation method for user is shown,
It can specifically include following steps:
Step 101, the incidence relation between two two users is established according to the characteristic information of user.
Characteristic information refers to the data information of the features such as each generic attribute of user, behavior, for example, the name of user, phone,
Mailbox, identification card number, ship-to, consignee's information of purchase order, connection network mark, subscribe viewing ticket when fill in
User personal information etc..Any suitable various features information is specifically included, the embodiment of the present invention is without limitation.
In embodiments of the present invention, analysis is compared to the characteristic information of all users, according to the characteristic information of user,
Establish the incidence relation between two two users.Incidence relation between two two users refers to two users with certain association, example
Such as, as user harvest people's cell-phone number for filling in shopping with the cell-phone number of another user is, then two users
Between have incidence relation.Any suitable incidence relation is specifically included, the embodiment of the present invention is without limitation.
For example, according to the characteristic information of all users, two two users with same characteristic features information are searched, it is dual-purpose to establish two
Incidence relation between family, wherein incidence relation specifically can be by the marks for recording two users and identical characteristic information.
Step 102, at least one user group is established in all users.
In embodiments of the present invention, any two user is directly linked or by between at least one user in user group
Connect association.After the incidence relation between two two users being established for all users, based on the incidence relation between two two users,
At least one user group is established in all users.All users can be specifically found out according to the incidence relation between user
Group can also find out number of users and reach a certain number of user groups, can specifically find out any number of user
Group, the embodiment of the present invention are without limitation.
In embodiments of the present invention, the user group established in all users may include all direct or indirect tool
Related user can also directly or indirectly have associated user, the embodiment of the present invention not to limit this including part
System.
Step 103, the feature user in the user group is determined.
In embodiments of the present invention, feature user refers to the certain customers selected from user group, specifically can basis
The quantity for the incidence relation established with other users determines feature user, or can be true according to the comment number or browsing number of user
Determine feature user, feature user is either determined according to the essential information of user or can be determined according to any suitable mode
Feature user, the embodiment of the present invention are without limitation.
It is characterized user for example, the user Jing Guo authentication is determined or will be used with other in each user group
User's determination that the incidence relation of family foundation is most is characterized user or is commented on or browsed the most use of number by picture and text message
Family, which determines, is characterized user.
Step 104, the recommendation of network object is carried out based on the feature user.
In embodiments of the present invention, network object refers to for the arbitrary service that can be accessed by network or object, clothes
Business can at least one of be ordered air ticket, order hotel, take-away etc., and object can be in commodity, picture, video, music, novel etc.
It is at least one.
In embodiments of the present invention, feature based user carries out the recommendation of network object, can be specifically by network object
The feature user in user group is recommended, or the network object that feature user accesses is recommended to the removing in user group
The other users of feature user specifically may be used any suitable mode feature based user and recommend, and the present invention is implemented
Example is without limitation.
In conclusion embodiment according to the present invention, by establishing the pass between two two users according to the characteristic information of user
Connection relationship establishes at least one user group in all users, determines the feature user in the user group, and be based on institute
State the recommendation that feature user carries out network object so that user can be divided into different groups according to the characteristic information of user
Body, and the user with incidence relation to the preference of network object with certain similitude, recommend for user group
Service can more recommend to service or commodity of the user really with demand, improve the efficiency of recommendation.
In one preferred embodiment of the invention, the characteristic information include user attributive character and behavioural characteristic in
It is at least one, wherein attributive character includes:Mark, title, contact method data;Behavioural characteristic includes:The target of access behavior
User, the attribute information of other users of submission, the network connected mark or area identification.
Attributive character refer to the relevant feature of user property, can be specifically the feature being closely related with user identity,
Including mark, title, contact method data etc..Behavioural characteristic refer to the relevant feature of the behavior of user in a network, including
The target user of access behavior, the attribute information of the other users of submission, the network connected mark or area identification.
Access behavior refers to the behaviors such as contact in the network platform, click, download, purchase, on probation, accesses the mesh of behavior
Mark user may include by correspondent party, be clicked the head portrait or the corresponding user of private page, the file for being downloaded upload of access
User, purchased side, by side on probation etc..The attribute information of the other users of submission refer to user submitted in the network platform appoint
The attribute informations such as mark, title, the contact method of other users filled in when information of anticipating.The mark of the network connected or region
Mark refers to the name identification or zone name of the WLAN connected, the address of the cable network connected or the network segment
Deng.
According to various features information, the incidence relation between various users can be found, as long as generating connection between user
System, so that it may to establish the incidence relation between user.
In one preferred embodiment of the invention, the step of determining the feature user in the user group include:Needle
To each user of the user group, statistics characterizes the associated data of the incidence relation of the user and other users, will close
The user of connection data fit preset requirement is determined as the feature user in the user group.
Associated data is the data of incidence relation between can characterizing user, for example, establishing the use of incidence relation with user
The quantity etc. that the picture and text message that the quantity at family, user deliver is commented on by other users, can specifically include any suitable association
Data, the embodiment of the present invention are without limitation.For each user group, statistical form levies each user and group in user group
The associated data of the incidence relation of other users in body.
For each user group, judge whether the associated data of each user meets preset requirement, associated data is accorded with
The user for closing preset requirement is determined as the feature user in user group.For example, by being built with other users in each user group
The most user of vertical incidence relation, which determines, is characterized user or to be commented on or browsed the most user of number by picture and text message true
It is set to feature user.
The user that associated data is met to preset requirement is determined as feature user in the user group, can find
With other users there are more associated feature users, feature user can be recommended network object more in user group
User, further diffusion to more users, recommend the efficiency that can improve recommendation for feature user, and overcome
Consumer is unwilling to believe the problem of service and commodity that platform is directly recommended, and network object recommended feature user, into
And feature user is recommended again in user group, other users are more willing to the content for receiving to recommend, to improve recommendation
Conversion ratio.
In one preferred embodiment of the invention, the associated data includes network behavior data, the network behavior
Data be characterize user and other users correlation degree data, associated data is met preset requirement user be determined as described in
The step of feature user in user group includes:In each user group, the network behavior data of each user are obtained;
Search the user that the network behavior data reach preset requirement, the feature user being determined as in the user group.
Associated data may include network behavior data, and network behavior data are characterization user and other users correlation degree
Data, for example, the picture and text message that user delivers by comment number, by browsing number etc., network behavior data can be one system
Evaluation, for example, the summation by comment number for all picture and text message that user delivers.
For each user group, the network behavior data of each user are obtained, the network behavior data is searched and reaches
The user of preset requirement, for example, preset requirement can be that be commented on number be more than 10,000, then the user found is user group
In body it is all by comment number be more than 10,000 user.Any suitable preset requirement, the present invention are used with specific reference to actual demand
Embodiment is without limitation.It finds in user group by this method, most powerful or transmission capacity user.It will
The user found is determined as the feature user in user group.
In one preferred embodiment of the invention, the step of carrying out the recommendation of network object based on the feature user is wrapped
It includes:Obtain all users for accessing network object;At least one user group is searched in all users for accessing network object
Feature user;The network object is recommended to the non-feature user in the user group of each feature user attaching.
It refers to buying, making a reservation for, clicking, downloading, using etc. in multiple networks for network object to access network object
Behavior accesses from the history for network object and obtains all users for accessing network object in data, can be specifically to obtain
The mark of all users.The feature user of at least one user group is searched in all users for accessing network object.Right
When network object is recommended, find in all users of network object be feature user user, network object is recommended
Non- feature user in the user group of each feature user attaching.Due to belonging to the user in a user group to network pair
The preference of elephant have similitude, and feature user in user group with more users have be associated with, feature user access
The network object crossed, network pair that feature user accessed stronger with the similitude of the preference of more users in user group
As recommending the non-feature user in user group, the conversion ratio of recommendation can be improved.
In one preferred embodiment of the invention, the network object includes network service, data object.Network service
Refer to the service provided by network, such as orders air ticket, order hotel, order the services such as take-away.Data object refers to being carried by network
The objects such as commodity, video, the document of confession, such as the toy of sale, collection of drama, novel etc..
With reference to Fig. 2, the step flow chart of a kind of according to embodiments of the present invention two recommendation method for user is shown,
It can specifically include following steps:
Step 201, the incidence relation between two two users is established according to the characteristic information of user.
In embodiments of the present invention, the specific implementation of this step may refer to the description of previous embodiment, herein not
Separately repeat.
Step 202, it is using each user as node, with the incidence relation of each user couple with same characteristic features information
Customer relationship figure is established on side.
In embodiments of the present invention, customer relationship figure refer to include all users, and connected with incidence relation between user
The relational graph of each user.Node and side are preserved in customer relationship figure, and specifically using each user as node, there is phase with each
Incidence relation with the user couple of characteristic information is side.Node in customer relationship figure can be preserved including user identifier and spy
The data such as reference breath, the side in customer relationship figure can preserve the mark and common feature letter of the user at a line both ends
Breath.
In one preferred embodiment of the invention, the step of establishing customer relationship figure include:Obtain each user identifier
And node data of the corresponding at least one characteristic information as each user;The user is obtained to corresponding user identifier
Number of edges evidence with common trait information as each user couple;It establishes by the node data and number of edges according to the customer relationship formed
Figure.
Node data refers to the data for including each user identifier and corresponding at least one characteristic information, and number of edges evidence is
Finger includes data of the user to corresponding user identifier and common trait information, then with node data and number of edges according to being constituted user
Relational graph.
Specifically, customer relationship figure can be established on diagram data server cluster, all nodes of cutting uniform first
Data are separately sent to preserve on each Analysis server of diagram data server cluster.According to the node data, tool is searched
The node pair for having same characteristic features information is believed according to the corresponding user identifier of the two of the node pair nodes and common trait
Breath, generates the number of edges evidence of each user couple.When specific implementation, each diagram data server can be called to search and store thereon respectively
Node data each node for including there is the destination node of same characteristic features information, the node data of the destination node can be with
It is stored on Subgraph data server.Each diagram data server is called to obtain two nodes of the node pair that it finds
The corresponding user identifier and common trait information.Each diagram data server is called to preserve the described of the node pair of its acquisition
The number of edges evidence of user identifier and common trait information as each user couple.
In the corresponding user identifier of two nodes and common trait information according to the node pair, each use is generated
The number of edges at family pair can also search all user identifiers having the same and be total in the diagram data server cluster after
With the number of edges evidence of characteristic information, the number of edges evidence of repetition is deleted, until the number of edges evidence preserved in the diagram data server cluster does not have
There is repetition.The number of edges evidence of cutting is distributed to the figure number by the number of edges evidence preserved on each diagram data server of uniform cutting
According to each diagram data server in server cluster, thus customer relationship figure is established in the completion of diagram data server cluster.
Step 203, the connected graph that continuous continual multiple nodes are constituted is searched from the customer relationship figure.
In embodiments of the present invention, the connection that continuous continual multiple nodes are constituted is searched from the customer relationship figure
Scheme, any two node can be connected to directly or by least one other node in connected graph.The connected graph specifically searched can
To be maximal connected graphs, maximal connected graphs are may not be, the embodiment of the present invention is without limitation.
In one preferred embodiment of the invention, the number of edges evidence of the customer relationship figure is distributed in diagram data server set
On group, any two diagram data server is directly linked or passes through at least one diagram data in the diagram data server cluster
The step of server indirect association, the connected graph that the continuous continual multiple nodes of lookup are constituted from the customer relationship figure, can
To include:It calls each diagram data server according to the number of edges evidence stored thereon, searches continuous continual multiple sections respectively
At least one sub- connected graph that point is constituted;Each diagram data server is called to be distributed to the sub- connected graph and each diagram data service
The associated at least one other diagram data server of device;It calls each diagram data server iteration to execute merging to find or receive
Sub- connected graph, and the sub- connected graph is distributed to and the associated at least one other diagram data of each diagram data server takes
Business device, until without annexable sub- connected graph.
Since number of users is huge, the data volume of customer relationship figure is too big, can not using single server processing diagram data
Meet demand can carry out the relevant operation of diagram data on diagram data server cluster.Specifically, the side of customer relationship figure
Data distribution on diagram data server cluster, in diagram data server cluster any two diagram data server be directly linked or
It is by least one diagram data server indirect association.
When specific implementation, each diagram data server is called, according to the number of edges evidence stored thereon, is searched respectively continuous uninterrupted
Multiple nodes constitute at least one sub- connected graph, wherein each sub- connected graph only has each independent basis of diagram data server
The number of edges evidence of its own storage obtains, and is not communicated with other diagram data servers, can be specifically if two number of edges evidences
Two end nodes, node having the same, then two sides belong to same sub- connected graph.Then, each diagram data server is called
By sub- connected graph be distributed to the associated at least one other diagram data server of each diagram data server, specifically by sub- connection
Number of edges evidence in figure and sub- connected graph is distributed to other diagram data servers.Each diagram data server receives other servers
After the sub- connected graph sent, according to the sub- connected graph of sub- connected graph and reception that oneself is searched, and merges, there will be same node point
Sub- connected graph be merged into one, be then distributed again, the son for merging and finding or receiving executed with this continuous iteration
Connected graph, and the sub- connected graph is distributed to and the associated at least one other diagram data service of each diagram data server
Device, until without annexable sub- connected graph.Since any two diagram data server directly closes in diagram data server cluster
Join or by least one diagram data server indirect association, after excessively taking turns iteration, diagram data server cluster is used
Connected graph in the relational graph of family.The work for searching connected graph is carried out parallel on different machines, as long as each
Result on machine summarize can, network overhead is also smaller, improves operational efficiency.
Step 204, using in connected graph the corresponding multiple users of all nodes as a user group.
In embodiments of the present invention, after obtaining connected graph, each node in connected graph is obtained, with all nodes in connected graph
Corresponding multiple users are as a user group.
Step 205, for each user of the user group, the statistics characterization user is associated with pass with other users
The associated data of system.
In embodiments of the present invention, the specific implementation of this step may refer to the description of previous embodiment, herein not
Separately repeat.
Step 206, the user for having incidence relation most with other users, the spy being determined as in the user group are searched
Take over family for use.
In embodiments of the present invention, associated data includes the quantity for the user for having incidence relation with user.For each
User group searches the user for wherein having incidence relation most with other users, and the user found is determined and is characterized use
Family.
Step 207, all users for accessing network object are obtained.
In embodiments of the present invention, for the network object to be recommended, all users for accessing network object, example are obtained
Such as, all characteristic informations for using the user for ordering ticketing service in platform are obtained.
Step 208, the user group for each user attaching for accessing network object is searched.
In embodiments of the present invention, according to each user group found, each user for accessing network object is searched
The user group of ownership.
Step 209, the network object is recommended to the feature user in the user group respectively found.
In embodiments of the present invention, network object is recommended to the feature user in the user group respectively found, for example,
This service recommendation of air ticket will be ordered to the feature user in user group.
In conclusion embodiment according to the present invention, by establishing the pass between two two users according to the characteristic information of user
Connection relationship establishes user using each user as node, using the incidence relation of each user couple with same characteristic features information as side
Relational graph searches the connected graph that continuous continual multiple nodes are constituted, to own in connected graph from the customer relationship figure
The corresponding multiple users of node are as a user group so that can establish the relationship of user according to the characteristic information of user
Figure, to find that each incidence relation between user, then the relational graph based on user divide user by searching for connected graph
For different groups, and the user with incidence relation to the preference of network object with certain similitude, for user group
Body carries out recommendation service, can more recommend to service or commodity of the user really with demand, improve the efficiency of recommendation.
Further, by each user for the user group, statistics characterizes the pass of the user and other users
The associated data of connection relationship is searched the user for having incidence relation most with other users, is determined as in the user group
Feature user obtains all users for accessing network object, searches the user for each user attaching for accessing network object
The network object is recommended the feature user in the user group respectively found by group so that can find and other
There are user more associated feature users, feature user network object can be recommended to more use in user group
Family, further diffusion recommend the efficiency that can improve recommendation for feature user, and overcome and disappear to more users
The problem of expense person is unwilling to believe the service directly recommended of platform and commodity, feature user, Jin Erte are recommended by network object
Requisition family is recommended again in user group, and other users are more willing to the content for receiving to recommend, to improve the conversion of recommendation
Rate.
It should be noted that for embodiment of the method, for simple description, therefore it is all expressed as a series of action group
It closes, but those skilled in the art should understand that, the embodiment of the present invention is not limited by the described action sequence, because according to
According to the embodiment of the present invention, certain steps can be performed in other orders or simultaneously.Secondly, those skilled in the art also should
Know, embodiment described in this description belongs to preferred embodiment, and the involved action not necessarily present invention is implemented
Necessary to example.
Embodiment three
With reference to Fig. 3, a kind of structure diagram of according to embodiments of the present invention three recommendation apparatus for user is shown, have
Body may include following module:
Relationship establishes module 301, for establishing the incidence relation between two two users according to the characteristic information of user;
Group establishes module 302, for establishing at least one user group in all users, appoints in the user group
Two users of meaning are directly linked or pass through at least one user's indirect association;
Feature user determining module 303, for determining the feature user in the user group;
Network object recommending module 304, the recommendation for carrying out network object based on the feature user.
In embodiments of the present invention, it is preferable that the characteristic information include user attributive character and behavioural characteristic in extremely
Few one kind, wherein attributive character includes:Mark, title, contact method data;Behavioural characteristic includes:The target of access behavior is used
Family, the attribute information of other users of submission, the network connected mark or area identification.
In embodiments of the present invention, it is preferable that the group establishes module and includes:
Relational graph setting up submodule is used for using each user as node, with each user couple with same characteristic features information
Incidence relation be side, establish customer relationship figure;
Connected graph searches submodule, for searching continuous continual multiple nodes compositions from the customer relationship figure
Connected graph;
Group's node submodule, for using in connected graph the corresponding multiple users of all nodes as a user group.
In embodiments of the present invention, it is preferable that the relational graph setting up submodule includes:
Node data acquiring unit, for obtaining each user identifier and corresponding at least one characteristic information as each
The node data of a user;
Side data capture unit, for obtaining the user to corresponding user identifier and common trait information as each
The number of edges evidence of user couple;
Relational graph establishes unit, for establishing by the node data and number of edges according to the customer relationship figure formed.
In embodiments of the present invention, it is preferable that the number of edges evidence of the customer relationship figure is distributed in diagram data server cluster
On, any two diagram data server is directly linked or is taken by least one diagram data in the diagram data server cluster
Business device indirect association, the connected graph search submodule and include:
Sub- connected graph searching unit, for calling each diagram data server according to the number of edges evidence stored thereon, respectively
Search at least one sub- connected graph that continuous continual multiple nodes are constituted;
Sub- connected graph Dispatching Unit, for calling each diagram data server to be distributed to the sub- connected graph and each diagram data
The associated at least one other diagram data server of server;
Merge and distribute iteration unit, for calling each diagram data server iteration to execute what merging found or received
Sub- connected graph, and the sub- connected graph is distributed to and the associated at least one other diagram data service of each diagram data server
Device, until without annexable sub- connected graph.
In embodiments of the present invention, it is preferable that the feature user determining module includes:
Associated data statistic submodule, for each user for the user group, statistics characterize the user with
The associated data of the incidence relation of other users;
Feature user's determination sub-module, the user for associated data to be met to preset requirement are determined as the user group
In feature user.
In embodiments of the present invention, it is preferable that the associated data includes the number for the user for having incidence relation with user
Amount, the feature user determination sub-module are specifically used for searching the user for having incidence relation most with other users, be determined as
Feature user in the user group.
In embodiments of the present invention, it is preferable that the associated data includes network behavior data, the network behavior data
To characterize the data of user and other users correlation degree, the feature user determination sub-module includes:
Behavioral data acquiring unit, in each user group, obtaining the network behavior data of each user;
Second feature user's determination unit reaches the user of preset requirement for searching the network behavior data, determines
For the feature user in the user group.
In embodiments of the present invention, it is preferable that the network object recommending module includes:
First user's acquisition submodule, for obtaining all users for accessing network object;
User group searches submodule, the user group for searching each user attaching for accessing network object;
First network object recommendation submodule, for recommending the network object in the user group respectively found
Feature user.
In embodiments of the present invention, it is preferable that the network object recommending module includes:
Second user acquisition submodule, for obtaining all users for accessing network object;
Feature user searches submodule, for searching at least one user group in all users for accessing network object
Feature user;
Second network object recommends submodule, the user group for the network object to be recommended to each feature user attaching
Non- feature user in body.
In embodiments of the present invention, it is preferable that the network object includes network service, data object.
In conclusion embodiment according to the present invention, by establishing the pass between two two users according to the characteristic information of user
Connection relationship establishes at least one user group in all users, determines the feature user in the user group, and be based on institute
State the recommendation that feature user carries out network object so that user can be divided into different groups according to the characteristic information of user
Body, and the user with incidence relation to the preference of network object with certain similitude, recommend for user group
Service can more recommend to service or commodity of the user really with demand, improve the efficiency of recommendation.
Example IV
A kind of computer readable storage medium is present embodiments provided, computer program is stored thereon with, which is characterized in that
The program realizes following steps when being executed by processor:
The incidence relation between two two users is established according to the characteristic information of user;
Establish at least one user group in all users, in the user group any two user be directly linked or
Pass through at least one user's indirect association;
Determine the feature user in the user group;
The recommendation of network object is carried out based on the feature user.
In embodiments of the present invention, it is preferable that the characteristic information include user attributive character and behavioural characteristic in extremely
Few one kind, wherein attributive character includes:Mark, title, contact method data;Behavioural characteristic includes:The target of access behavior is used
Family, the attribute information of other users of submission, the network connected mark or area identification.
In embodiments of the present invention, it is preferable that described the step of establishing at least one user group in all users wraps
It includes:
Using each user as node, using the incidence relation of each user couple with same characteristic features information as side, establish use
Family relational graph;
The connected graph that continuous continual multiple nodes are constituted is searched from the customer relationship figure;
The corresponding multiple users of all nodes are as a user group using in connected graph.
In embodiments of the present invention, it is preferable that described the step of establishing customer relationship figure includes:
Obtain the node data of each user identifier and corresponding at least one characteristic information as each user;
Obtain number of edges evidence of the user to corresponding user identifier and common trait information as each user couple;
It establishes by the node data and number of edges according to the customer relationship figure formed.
In embodiments of the present invention, it is preferable that the number of edges evidence of the customer relationship figure is distributed in diagram data server cluster
On, any two diagram data server is directly linked or is taken by least one diagram data in the diagram data server cluster
Business device indirect association, the described the step of connected graph that continuous continual multiple nodes are constituted is searched from the customer relationship figure
Including:
It calls each diagram data server according to the number of edges evidence stored thereon, searches continuous continual multiple sections respectively
At least one sub- connected graph that point is constituted;
Each diagram data server is called to be distributed to the sub- connected graph associated at least one with each diagram data server
Other diagram data servers;
Each diagram data server iteration is called to execute the sub- connected graph for merging and finding or receiving, and even by the son
Logical figure be distributed to the associated at least one other diagram data server of each diagram data server, until it is annexable son even
Logical figure.
In embodiments of the present invention, it is preferable that the step of feature user in the determination user group includes:
For each user of the user group, the statistics characterization user is associated with the incidence relation of other users
Data;
The user that associated data is met to preset requirement is determined as feature user in the user group.
In embodiments of the present invention, it is preferable that the associated data includes the number for the user for having incidence relation with user
Amount, the step of feature user that the user that associated data is met to preset requirement is determined as in the user group include:
Search the user for having incidence relation most with other users, the feature user being determined as in the user group.
In embodiments of the present invention, it is preferable that the associated data includes network behavior data, the network behavior data
To characterize the data of user and other users correlation degree, the user that associated data is met to preset requirement is determined as described
The step of feature user in user group includes:
In each user group, the network behavior data of each user are obtained;
The user that the network behavior data reach preset requirement is searched, the feature being determined as in the user group is used
Family.
In embodiments of the present invention, it is preferable that it is described based on the feature user carry out network object recommendation the step of
Including:
Obtain all users for accessing network object;
Search the user group for each user attaching for accessing network object;
The network object is recommended to the feature user in the user group respectively found.
In embodiments of the present invention, it is preferable that it is described based on the feature user carry out network object recommendation the step of
Including:
Obtain all users for accessing network object;
The feature user of at least one user group is searched in all users for accessing network object;
The network object is recommended to the non-feature user in the user group of each feature user attaching.
In embodiments of the present invention, it is preferable that the network object includes network service, data object.
In conclusion embodiment according to the present invention, by establishing the pass between two two users according to the characteristic information of user
Connection relationship establishes at least one user group in all users, determines the feature user in the user group, and be based on institute
State the recommendation that feature user carries out network object so that user can be divided into different groups according to the characteristic information of user
Body, and the user with incidence relation to the preference of network object with certain similitude, recommend for user group
Service can more recommend to service or commodity of the user really with demand, improve the efficiency of recommendation.
Embodiment five
Present embodiments provide a kind of computer equipment, including memory, processor and storage are on a memory and can be
The computer program run on processor, which is characterized in that the processor realizes following steps when executing described program:
The incidence relation between two two users is established according to the characteristic information of user;
Establish at least one user group in all users, in the user group any two user be directly linked or
Pass through at least one user's indirect association;
Determine the feature user in the user group;
The recommendation of network object is carried out based on the feature user.
In embodiments of the present invention, it is preferable that the characteristic information include user attributive character and behavioural characteristic in extremely
Few one kind, wherein attributive character includes:Mark, title, contact method data;Behavioural characteristic includes:The target of access behavior is used
Family, the attribute information of other users of submission, the network connected mark or area identification.
In embodiments of the present invention, it is preferable that described the step of establishing at least one user group in all users wraps
It includes:
Using each user as node, using the incidence relation of each user couple with same characteristic features information as side, establish use
Family relational graph;
The connected graph that continuous continual multiple nodes are constituted is searched from the customer relationship figure;
The corresponding multiple users of all nodes are as a user group using in connected graph.
In embodiments of the present invention, it is preferable that described the step of establishing customer relationship figure includes:
Obtain the node data of each user identifier and corresponding at least one characteristic information as each user;
Obtain number of edges evidence of the user to corresponding user identifier and common trait information as each user couple;
It establishes by the node data and number of edges according to the customer relationship figure formed.
In embodiments of the present invention, it is preferable that the number of edges evidence of the customer relationship figure is distributed in diagram data server cluster
On, any two diagram data server is directly linked or is taken by least one diagram data in the diagram data server cluster
Business device indirect association, the described the step of connected graph that continuous continual multiple nodes are constituted is searched from the customer relationship figure
Including:
It calls each diagram data server according to the number of edges evidence stored thereon, searches continuous continual multiple sections respectively
At least one sub- connected graph that point is constituted;
Each diagram data server is called to be distributed to the sub- connected graph associated at least one with each diagram data server
Other diagram data servers;
Each diagram data server iteration is called to execute the sub- connected graph for merging and finding or receiving, and even by the son
Logical figure be distributed to the associated at least one other diagram data server of each diagram data server, until it is annexable son even
Logical figure.
In embodiments of the present invention, it is preferable that the step of feature user in the determination user group includes:
For each user of the user group, the statistics characterization user is associated with the incidence relation of other users
Data;
The user that associated data is met to preset requirement is determined as feature user in the user group.
In embodiments of the present invention, it is preferable that the associated data includes the number for the user for having incidence relation with user
Amount, the step of feature user that the user that associated data is met to preset requirement is determined as in the user group include:
Search the user for having incidence relation most with other users, the feature user being determined as in the user group.
In embodiments of the present invention, it is preferable that the associated data includes network behavior data, the network behavior data
To characterize the data of user and other users correlation degree, the user that associated data is met to preset requirement is determined as described
The step of feature user in user group includes:
In each user group, the network behavior data of each user are obtained;
The user that the network behavior data reach preset requirement is searched, the feature being determined as in the user group is used
Family.
In embodiments of the present invention, it is preferable that it is described based on the feature user carry out network object recommendation the step of
Including:
Obtain all users for accessing network object;
Search the user group for each user attaching for accessing network object;
The network object is recommended to the feature user in the user group respectively found.
In embodiments of the present invention, it is preferable that it is described based on the feature user carry out network object recommendation the step of
Including:
Obtain all users for accessing network object;
The feature user of at least one user group is searched in all users for accessing network object;
The network object is recommended to the non-feature user in the user group of each feature user attaching.
In embodiments of the present invention, it is preferable that the network object includes network service, data object.
In conclusion embodiment according to the present invention, by establishing the pass between two two users according to the characteristic information of user
Connection relationship establishes at least one user group in all users, determines the feature user in the user group, and be based on institute
State the recommendation that feature user carries out network object so that user can be divided into different groups according to the characteristic information of user
Body, and the user with incidence relation to the preference of network object with certain similitude, recommend for user group
Service can more recommend to service or commodity of the user really with demand, improve the efficiency of recommendation.
For device embodiments, since it is basically similar to the method embodiment, so fairly simple, the correlation of description
Place illustrates referring to the part of embodiment of the method.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with
The difference of other embodiment, the same or similar parts between the embodiments can be referred to each other.
It should be understood by those skilled in the art that, the embodiment of the embodiment of the present invention can be provided as method, apparatus or calculate
Machine program product.Therefore, the embodiment of the present invention can be used complete hardware embodiment, complete software embodiment or combine software and
The form of the embodiment of hardware aspect.Moreover, the embodiment of the present invention can be used one or more wherein include computer can
With in the computer-usable storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.) of program code
The form of the computer program product of implementation.
The embodiment of the present invention be with reference to according to the method for the embodiment of the present invention, terminal device (system) and computer program
The flowchart and/or the block diagram of product describes.It should be understood that flowchart and/or the block diagram can be realized by computer program instructions
In each flow and/or block and flowchart and/or the block diagram in flow and/or box combination.These can be provided
Computer program instructions are set to all-purpose computer, special purpose computer, Embedded Processor or other programmable data processing terminals
Standby processor is to generate a machine so that is held by the processor of computer or other programmable data processing terminal equipments
Capable instruction generates for realizing in one flow of flow chart or multiple flows and/or one box of block diagram or multiple boxes
The device of specified function.
These computer program instructions, which may also be stored in, can guide computer or other programmable data processing terminal equipments
In computer-readable memory operate in a specific manner so that instruction stored in the computer readable memory generates packet
The manufacture of command device is included, which realizes in one flow of flow chart or multiple flows and/or one side of block diagram
The function of being specified in frame or multiple boxes.
These computer program instructions can be also loaded into computer or other programmable data processing terminal equipments so that
Series of operation steps are executed on computer or other programmable terminal equipments to generate computer implemented processing, thus
The instruction executed on computer or other programmable terminal equipments is provided for realizing in one flow of flow chart or multiple flows
And/or in one box of block diagram or multiple boxes specify function the step of.
Although the preferred embodiment of the embodiment of the present invention has been described, once a person skilled in the art knows bases
This creative concept, then additional changes and modifications can be made to these embodiments.So the following claims are intended to be interpreted as
Including preferred embodiment and fall into all change and modification of range of embodiment of the invention.
Finally, it is to be noted that, herein, relational terms such as first and second and the like be used merely to by
One entity or operation are distinguished with another entity or operation, without necessarily requiring or implying these entities or operation
Between there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant meaning
Covering non-exclusive inclusion, so that process, method, article or terminal device including a series of elements not only wrap
Those elements are included, but also include other elements that are not explicitly listed, or further include for this process, method, article
Or the element that terminal device is intrinsic.In the absence of more restrictions, being wanted by what sentence "including a ..." limited
Element, it is not excluded that there is also other identical elements in process, method, article or the terminal device including the element.
Above to a kind of recommendation method, apparatus, equipment and storage medium for user provided by the present invention, carry out
It is discussed in detail, principle and implementation of the present invention are described for specific case used herein, above example
Illustrate the method and its core concept for being merely used to help understand the present invention;Meanwhile for those of ordinary skill in the art, according to
According to the thought of the present invention, there will be changes in the specific implementation manner and application range, in conclusion the content of the present specification
It should not be construed as limiting the invention.
Claims (14)
1. a kind of recommendation method for user, which is characterized in that including:
The incidence relation between two two users is established according to the characteristic information of user;
At least one user group is established in all users, any two user is directly linked or passes through in the user group
At least one user's indirect association;
Determine the feature user in the user group;
The recommendation of network object is carried out based on the feature user.
2. according to the method described in claim 1, it is characterized in that, the characteristic information includes attributive character and the behavior of user
At least one of feature, wherein attributive character includes:Mark, title, contact method data;Behavioural characteristic includes:Access behavior
Target user, the attribute information of other users of submission, the network connected mark or area identification.
3. according to the method described in claim 1, it is characterized in that, described establish at least one user group in all users
The step of include:
Using each user as node, using the incidence relation of each user couple with same characteristic features information as side, establish user pass
System's figure;
The connected graph that continuous continual multiple nodes are constituted is searched from the customer relationship figure;
The corresponding multiple users of all nodes are as a user group using in connected graph.
4. according to the method described in claim 3, it is characterized in that, described the step of establishing customer relationship figure include:
Obtain the node data of each user identifier and corresponding at least one characteristic information as each user;
Obtain number of edges evidence of the user to corresponding user identifier and common trait information as each user couple;
It establishes by the node data and number of edges according to the customer relationship figure formed.
5. according to the method described in claim 4, it is characterized in that, the number of edges of the customer relationship figure takes according to diagram data is distributed in
It is engaged on device cluster, any two diagram data server is directly linked or by least one in the diagram data server cluster
Diagram data server indirect association, the connection that continuous continual multiple nodes are searched from the customer relationship figure and are constituted
The step of figure includes:
It calls each diagram data server according to the number of edges evidence stored thereon, searches continuous continual multiple node structures respectively
At at least one sub- connected graph;
Each diagram data server is called to be distributed to the sub- connected graph associated at least one other with each diagram data server
Diagram data server;
Each diagram data server iteration is called to execute the sub- connected graph for merging and finding or receiving, and by the sub- connected graph
Be distributed to the associated at least one other diagram data server of each diagram data server, until without annexable sub- connection
Figure.
6. according to the method described in claim 1, it is characterized in that, the step of the feature user in the determination user group
Suddenly include:
For each user of the user group, statistics characterizes the incidence number of the incidence relation of the user and other users
According to;
The user that associated data is met to preset requirement is determined as feature user in the user group.
7. according to the method described in claim 6, it is characterized in that, the associated data includes having incidence relation with user
The quantity of user, the user that associated data is met to preset requirement are determined as the step of the feature user in the user group
Suddenly include:
Search the user for having incidence relation most with other users, the feature user being determined as in the user group.
8. according to the method described in claim 6, it is characterized in that, the associated data includes network behavior data, the net
Network behavioral data is the data for characterizing user and other users correlation degree, the user that associated data is met to preset requirement
The step of feature user being determined as in the user group includes:
In each user group, the network behavior data of each user are obtained;
Search the user that the network behavior data reach preset requirement, the feature user being determined as in the user group.
9. according to the method described in claim 1, it is characterized in that, described carry out pushing away for network object based on the feature user
The step of recommending include:
Obtain all users for accessing network object;
Search the user group for each user attaching for accessing network object;
The network object is recommended to the feature user in the user group respectively found.
10. according to the method described in claim 1, it is characterized in that, described carry out network object based on the feature user
The step of recommendation includes:
Obtain all users for accessing network object;
The feature user of at least one user group is searched in all users for accessing network object;
The network object is recommended to the non-feature user in the user group of each feature user attaching.
11. according to the method described in claim 1, it is characterized in that, the network object includes network service, data object.
12. a kind of recommendation apparatus for user, which is characterized in that including:
Relationship establishes module, for establishing the incidence relation between two two users according to the characteristic information of user;
Group establishes module, for establishing at least one user group in all users, any two in the user group
User is directly linked or passes through at least one user's indirect association;
Feature user's determining module, for determining the feature user in the user group;
Network object recommending module, the recommendation for carrying out network object based on the feature user.
13. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is by processor
Any one of them method such as claim 1~11 is realized when execution.
14. a kind of computer equipment, including memory, processor and storage are on a memory and the meter that can run on a processor
Calculation machine program, which is characterized in that the processor realizes any one of them such as claim 1~11 when executing described program
Method.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810054377.3A CN108416645B (en) | 2018-01-19 | 2018-01-19 | Recommendation method, device, storage medium and equipment for user |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810054377.3A CN108416645B (en) | 2018-01-19 | 2018-01-19 | Recommendation method, device, storage medium and equipment for user |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108416645A true CN108416645A (en) | 2018-08-17 |
CN108416645B CN108416645B (en) | 2021-08-20 |
Family
ID=63125865
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810054377.3A Active CN108416645B (en) | 2018-01-19 | 2018-01-19 | Recommendation method, device, storage medium and equipment for user |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108416645B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110765358A (en) * | 2019-10-25 | 2020-02-07 | 秒针信息技术有限公司 | Information pushing method and device, electronic equipment and readable storage medium |
CN111277856A (en) * | 2020-02-24 | 2020-06-12 | 拉扎斯网络科技(上海)有限公司 | Information processing method and system |
CN111581506A (en) * | 2020-04-28 | 2020-08-25 | 海南太美航空股份有限公司 | Flight recommendation method and system based on collaborative filtering |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102063433A (en) * | 2009-11-16 | 2011-05-18 | 华为技术有限公司 | Method and device for recommending related items |
US20150120721A1 (en) * | 2013-10-25 | 2015-04-30 | Marketwire L.P. | Systems and Methods for Dynamically Determining Influencers in a Social Data Network Using Weighted Analysis |
CN105117443A (en) * | 2015-08-12 | 2015-12-02 | 华南理工大学 | Multi-relational network-based recommendation algorithm |
CN106446078A (en) * | 2016-09-08 | 2017-02-22 | 乐视控股(北京)有限公司 | Information recommendation method and recommendation apparatus |
CN106708844A (en) * | 2015-11-12 | 2017-05-24 | 阿里巴巴集团控股有限公司 | User group partitioning method and device |
-
2018
- 2018-01-19 CN CN201810054377.3A patent/CN108416645B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102063433A (en) * | 2009-11-16 | 2011-05-18 | 华为技术有限公司 | Method and device for recommending related items |
US20150120721A1 (en) * | 2013-10-25 | 2015-04-30 | Marketwire L.P. | Systems and Methods for Dynamically Determining Influencers in a Social Data Network Using Weighted Analysis |
CN105117443A (en) * | 2015-08-12 | 2015-12-02 | 华南理工大学 | Multi-relational network-based recommendation algorithm |
CN106708844A (en) * | 2015-11-12 | 2017-05-24 | 阿里巴巴集团控股有限公司 | User group partitioning method and device |
CN106446078A (en) * | 2016-09-08 | 2017-02-22 | 乐视控股(北京)有限公司 | Information recommendation method and recommendation apparatus |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110765358A (en) * | 2019-10-25 | 2020-02-07 | 秒针信息技术有限公司 | Information pushing method and device, electronic equipment and readable storage medium |
CN110765358B (en) * | 2019-10-25 | 2022-08-09 | 秒针信息技术有限公司 | Information pushing method and device, electronic equipment and readable storage medium |
CN111277856A (en) * | 2020-02-24 | 2020-06-12 | 拉扎斯网络科技(上海)有限公司 | Information processing method and system |
CN111581506A (en) * | 2020-04-28 | 2020-08-25 | 海南太美航空股份有限公司 | Flight recommendation method and system based on collaborative filtering |
Also Published As
Publication number | Publication date |
---|---|
CN108416645B (en) | 2021-08-20 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108960975A (en) | Personalized Precision Marketing Method, server and storage medium based on user's portrait | |
US9348916B2 (en) | Method and system for providing search services for a social media ecosystem | |
US9824404B2 (en) | Method and system for providing a social media ecosystem cooperative marketplace | |
Kim et al. | Personalized recommendation over a customer network for ubiquitous shopping | |
Costa-Montenegro et al. | Which App? A recommender system of applications in markets: Implementation of the service for monitoring users’ interaction | |
CN102226916B (en) | Platform, the method and apparatus of management application program | |
CN106168971A (en) | information subscribing method and device | |
US8843463B2 (en) | Providing content by using a social network | |
CN104077415B (en) | Searching method and device | |
JP5468076B2 (en) | Method and system for providing product object information on the Internet | |
US20080052371A1 (en) | System, apparatus and method for discovery of music within a social network | |
US20150066689A1 (en) | Method and system for providing social media ecosystem compensation | |
JP2001222601A (en) | System and method for information communication and information providing business method | |
CN109977296B (en) | Information pushing method, device, equipment and storage medium | |
CN106202115A (en) | A kind of orientation display information approach and device | |
CN109033149A (en) | Information recommendation method, device, server and storage medium | |
CN108416645A (en) | A kind of recommendation method, apparatus, storage medium and equipment for user | |
US20150294020A1 (en) | System and/or method for evaluating network content | |
CN101025740A (en) | Automatic play method of picture search result | |
CN109978580A (en) | Object recommendation method, apparatus and computer readable storage medium | |
JP2004220074A (en) | Net shopping method and program | |
CN110457610A (en) | Information recommendation method, device, terminal, server and storage medium | |
CN106779791A (en) | A kind of generation method and device of object picture combination of arranging in pairs or groups | |
CN106156351B (en) | Multimedia resource recommendation information generation method and device | |
CN104156375B (en) | A kind of information distribution method and system based on crowdsourcing |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |