CN108805642A - Recommend method and device - Google Patents
Recommend method and device Download PDFInfo
- Publication number
- CN108805642A CN108805642A CN201710300852.6A CN201710300852A CN108805642A CN 108805642 A CN108805642 A CN 108805642A CN 201710300852 A CN201710300852 A CN 201710300852A CN 108805642 A CN108805642 A CN 108805642A
- Authority
- CN
- China
- Prior art keywords
- node
- resource allocation
- resource
- collection
- initial
- 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.)
- Pending
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
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
This disclosure relates to recommend method and device, this method includes：Complete tripartite graph G (V are built based on user data, E), wherein, V=(U, I, T), U is that user node integrates, I integrates as item nodes, T is label node collection, and E is the initial connection set of relationship between user node collection, item nodes collection and label node concentration any two set of node；Resource allocation, which is carried out, according to initial resource and initial connection set of relationship obtains resource allocation result；It generates recommendation results according to resource allocation result and the recommendation results is sent to terminal device and show.By introducing label, the accuracy of recommendation can be improved according to the above-mentioned recommendation method and apparatus of the disclosure, realizes personalized recommendation, improves user experience.
Description
Technical field
This disclosure relates to applications of computer network technical field more particularly to a kind of recommendation method and device.
Background technology
Easily information is transmitted caused by network and information service has expedited the emergence of a large amount of network information.In face of so numerous
Information, user can not quick obtaining oneself need content, how using these data with improve operation, advertisement launch efficiency
It is of interest by more and more industries according to the content that the behavior of user, interest recommended user need as the hot spot of research.
In the related technology, the proposed algorithm for being based on bipartite graph (user, project (item)) is widely used in internet neck
Domain recommends it may interested information to user.Based on the proposed algorithm of bipartite graph to the excavation range of mass data information
It is limited with depth, the demand of user cannot be better met.
Invention content
In view of this, the present disclosure proposes a kind of recommendation method, the accuracy of recommendation is improved, personalized recommendation is realized, carries
High user experience.
According to the one side of the disclosure, a kind of recommendation method is provided, including：Complete tripartite graph is built based on user data
G (V, E), wherein V=(U, I, T), U are that user node integrates, I integrates as item nodes, T is label node collection, and E is user node
Collection, item nodes collection and label node concentrate the initial connection set of relationship between any two set of node；According to initial resource
And initial connection set of relationship carries out resource allocation and obtains resource allocation result；Recommendation results are generated according to resource allocation result
And the recommendation results are sent to terminal device and are shown.
According to another aspect of the present disclosure, a kind of recommendation apparatus is provided, including：Module is built, for being based on number of users
According to the complete tripartite graph G (V, E) of structure, wherein V=(U, I, T), U are that user node integrates, I integrates as item nodes, T is label section
Point set, E are the initial connection set of relations between user node collection, item nodes collection and label node concentration any two set of node
It closes；Resource distribution module obtains resource allocation for carrying out resource allocation according to initial resource and initial connection set of relationship
As a result；Recommending module generates recommendation results according to resource allocation result and the recommendation results is sent to terminal device progress
Display.
According to another aspect of the present disclosure, a kind of recommendation apparatus is provided, including：Processor；It can for storing processor
The memory executed instruction；Wherein, the processor is configured as executing the above method.
According to another aspect of the present disclosure, a kind of non-volatile computer readable storage medium storing program for executing is provided, is stored thereon with
Computer program instructions, which is characterized in that the computer program instructions realize the above method when being executed by processor.
By introducing label, complete tripartite graph is built based on user data and remains initial connection between all sets of node
Relationship, and resource allocation is carried out according to the initial connection relationship and the initial resource of distribution and obtains resource allocation result, profit
It is generated recommendation results with the resource allocation result and the recommendation results is sent to terminal device and shown.According to this public affairs
The accuracy of recommendation can be improved by opening above-mentioned recommendation method and apparatus, realize personalized recommendation, improve user experience.
According to below with reference to the accompanying drawings to detailed description of illustrative embodiments, the other feature and aspect of the disclosure will become
It is clear.
Description of the drawings
Including in the description and the attached drawing of a part for constitution instruction and specification together illustrate the disclosure
Exemplary embodiment, feature and aspect, and for explaining the principles of this disclosure.
Fig. 1 shows the flow chart of the recommendation method according to one embodiment of the disclosure.
Fig. 2 shows the intermediate node connection relationship diagrams according to the exemplary each set of node of the disclosure one.
Fig. 3 a- Fig. 3 d are shown according to one exemplary resource allocation schematic diagram of the disclosure.
Fig. 4 shows the flow chart of the recommendation method according to one embodiment of the disclosure.
Fig. 5 shows the flow chart of the recommendation method according to one embodiment of the disclosure.
Fig. 6 shows the flow chart according to one exemplary recommendation method of the disclosure.
Fig. 7 shows the block diagram of the recommendation apparatus according to one embodiment of the disclosure.
Fig. 8 shows the block diagram of the recommendation apparatus according to one embodiment of the disclosure.
Fig. 9 shows the block diagram of the recommendation apparatus according to one embodiment of the disclosure.
Specific implementation mode
Various exemplary embodiments, feature and the aspect of the disclosure are described in detail below with reference to attached drawing.It is identical in attached drawing
Reference numeral indicate functionally the same or similar element.Although the various aspects of embodiment are shown in the accompanying drawings, remove
It non-specifically points out, it is not necessary to attached drawing drawn to scale.
Dedicated word " exemplary " means " being used as example, embodiment or illustrative " herein.Here as " exemplary "
Illustrated any embodiment should not necessarily be construed as preferred or advantageous over other embodiments.
In addition, in order to better illustrate the disclosure, numerous details is given in specific implementation mode below.
It will be appreciated by those skilled in the art that without certain details, the disclosure can equally be implemented.In some instances, for
Method, means, element and circuit well known to those skilled in the art are not described in detail, in order to highlight the purport of the disclosure.
Embodiment 1
Fig. 1 shows that the flow chart of the recommendation method according to one embodiment of the disclosure, this method can be applied to server etc..
As shown in Figure 1, this method includes：
Step S11 builds complete tripartite graph G (V, E), wherein V=(U, I, T), U are user node based on user data
Integrate, I integrates as item nodes, T is label node collection, E is that user node collection, item nodes collection and label node concentrate any two
Initial connection set of relationship between set of node.
Server can obtain user data by terminal, and the user data may include user node collection, project section
Connection relation between point set, label node collection and any two set of node.It is used for example, server can be obtained by terminal
The node that the account information at family is concentrated as user node by article that user chose, the video watched or was listened
The node that music etc. is concentrated as item nodes, by label that user adds project (for example, user's " doing to video labeling
Laugh at ", " peace and quiet " etc. that music is marked) or item attribute make label possessed by project (for example, in micro- loudspeaker jeans
" micro- loudspeaker ") etc. as label node concentrate node；Server can also be obtained by terminal between any two set of node
Connection relation, by taking video as an example, connection relation may include that user " watched " certain video, user marks to certain video " addition "
Label or certain video " carry " label etc..
Server can build complete tripartite graph G (V, E) based on the user data of acquisition, wherein V=(U, I, T), U are
User node integrates, I integrates as item nodes, T is label node collection, and E is that user node collection, item nodes collection and label node are concentrated
Initial connection set of relationship between any two set of node.For example, U={ U1, U2... Un, I={ I1, I2... Im, T=
{T1, T2... Tp, wherein U1, U2 ... Un is user node, I1, I2... ImIt is item nodes, T1, T2... TpIt is label
Node.
In a kind of possible embodiment, the initial connection set of relationship includes：Set AUI, AITAnd AUT.Set AUI
The initial connection set of relationship between user node collection and item nodes collection is indicated, for example, user UiSelected project Ij, then collect
Close AUIIn corresponding element aij=1, otherwise, aij=0；Set AITIndicate initial between item nodes collection and label node collection
Connection relation set, for example, user is to project IiAdded label Tj, then set AITIn corresponding element bij=1, otherwise, bij
=0；Set AUTThe initial connection set of relationship between user node collection and label node collection is indicated, for example, user UiTo project
Added label Tj, then set AUTIn corresponding element cij=1, otherwise, cij=0.
Fig. 2 shows the intermediate node connection relationship diagrams according to the exemplary each set of node of the disclosure one.As shown in Fig. 2, three
Project in portion's figure can be specially article, and connection relation set can be indicated by the side of connection, such as：
Step S12 carries out resource allocation according to initial resource and initial connection set of relationship and obtains resource allocation result.
The initial resource can refer to the set of node of the starting point of diffusion or conduction have any can represent it
The resource of recommendation ability, it is understood that for energy or heat etc., for example, for a certain specific user, user buys
Certain commodity is crossed, then the corresponding resource of the commodity can be assigned as 1, and user did not bought certain commodity, then the corresponding resource of the commodity
It can be assigned as 0.Broadcast algorithm or heat transfer algorithm etc. in the related technology may be used, according to initial resource and initially
Connection relation set carries out resource allocation and obtains resource allocation result.
Fig. 3 a- Fig. 3 d are shown
Example illustrates, and is thought of as user (for example, shopper) U1Recommendations, for example, in article-user-label-article direction.
The user bought commodity I1、I3And I5, then, article I1-I5Initial resource it is as shown in Figure 3a.
For diffusion process, first in article-user direction, the resource of oneself is averagely given all purchases by each commodity
Bought its user, the resource of user be then from the summation of the obtained resource of all commodity, as shown in Figure 3b, user U1Money
Source is 3/2, user U2Resource be 1/2, user U3Resource be 1.
In user-tag orientation, the resource of oneself is averagely given all labels marked by each user, label
Resource be then from the summation of the obtained resource of all users, as shown in Figure 3c, label T1Resource be 3/4, label T2Resource
It is 1/2, label T3Resource be 3/2, label T4Resource be 1/4.
In label-article direction, the resource of oneself is averagely given all articles with the label, object by each label
The resource of product be then from the summation of the obtained resource of all labels, as shown in Figure 3d, article I1Resource be 5/8, article I2's
Resource is 7/8, article I3Resource be 3/8, article I4Resource be 5/8, article I5Resource be 1/2.
The result of resource allocation can be obtained according to above resource allocation process.It should be noted that although with shopping
It is as above that process, broadcast algorithm and article-user-label-article direction as example describe recommendation method, but this field skill
Art personnel it is understood that the disclosure answer it is without being limited thereto.In fact, user completely can be according to personal like and/or practical application field
Scape flexibly set resource allocation algorithm, distribution direction etc., the disclosure is not construed as limiting this.
Step S13, according to resource allocation result generate recommendation results and by the recommendation results be sent to terminal device into
Row display.
Still taking the above example as an example, can according to the corresponding resource value of all items in resource allocation result to article into
Row sequence, to user U1Recommend the forward one or several articles that sort, the article of recommendation is sent to user U1Terminal carry out
It has been shown that, can be shown in a manner of list.Such as it can be to user U1Recommend article I2, article I4Deng by article I2And I4's
Information is sent to user U in a manner of list1Terminal shown.
Compared to the proposed algorithm in the related technology based on bipartite graph, disclosure above-described embodiment by introducing label, and
Complete tripartite graph is built based on user data and remains initial connection relationship between all sets of node, and is initially connected according to described
The initial resource for connecing relationship and distribution carries out resource allocation acquisition resource allocation result, and the resource allocation result is utilized to generate
The recommendation results are simultaneously sent to terminal device and shown by recommendation results.It can be improved according to the above-mentioned recommendation method of the disclosure
The accuracy of recommendation realizes personalized recommendation, improves user experience.
Label is a kind of no hierarchical structure, for description information, shows the keyword of article semanteme, in tag system
In, label is acted on by user collaborative and being generated, and is the embodiment of group wisdom, simultaneously, label can promote similar users group
Generation, for using the user of same label, their interest preference tends to be close, and being introduced into personalized recommendation method has
Conducive to better description and analysis user interest, to preferably carry out personalized recommendation.
Fig. 4 shows the flow chart of the recommendation method according to one embodiment of the disclosure, as shown in figure 4, this method further includes：
Step S14, the time and number occurred according to connection relation determine the weight value set W between each set of node；
Step S15 determines weighting connection relation set according to the weight value set W and initial connection set of relationship.
The time and number occurred according to connection relation determines the weight value set between each set of node, and according to weighted value
Set and initial connection set of relationship obtain weighting connection relation set, and weighting connection relation set reflects user interest at any time
Between variation so that the result currently recommended have more reference value.It can be with according to the recommendation method of disclosure above-described embodiment
What dynamic adjustment was recommended recommends accuracy as a result, improving, and improves user experience.
The interest of user is not unalterable, and when user interest changes, the relevant technologies cannot react in time,
Cause recommendation effect bad, cannot be satisfied user demand.
Personalized recommendation is carried out to user, user interest preference information must just be captured, the interest of user
Preference can generate variation as time goes by, if user interest preference can not be reflected user's as if immobilizing
True interest.Therefore, the dynamic migration of user interest is added in a model, more suitably can accurately will recommend use by article
Family.The embodiment of time effect is, for example, the label that user adds recently can reflect its content being currently most interested in, and object
The label that product are added recently can also more properly reflect its content.Therefore, time of origin can be made apart from current time
Closer to influence bigger of the connection relation to weighted value.
In a kind of possible embodiment, the variation of user interest can be determined according to the time that connection relation occurs,
For example, the recommendation value that user has behavior, closer user behavior can be distinguished by connection relation time of origin
It indicates user's interested content recently, more recommendation values, more early user behavior can be contributed to indicate to close before user
It noted but less interested content recently, less recommendation is contributed to be worth.In addition, due to the number of connection relation generation
User interest can be reacted, when connection relation occur number more than once when, can also integrate connection relation generation time
The variation that user interest is determined with number is weighted the initial connection set of relationship in complete tripartite graph and is connected with obtaining weighting
Connect set of relationship.
Specifically, the time and number that can be occurred according to connection relation determine the weight value set W between each set of node,
Weight value set W may include：WUI、WITAnd WUT.Wherein, WUIIndicate the weighted value between user node collection and item nodes collection
Set, WITIndicate the weight value set between item nodes collection and label node collection, WUTIndicate user node collection and label node
Weight value set between collection.
In a kind of possible embodiment, the weighted value collection between each set of node can be determined according to following formula (1)
It closes：
Wherein, wijFor the weighted value between i-th of node of first segment point set and j-th of node of second node collection, t is
Current time, k are the number that connection relation occurs, tij,sIt is j-th of i-th node and the second node collection of first segment point set
The time that the s times connection relation of node occurs, w0Indicate weight threshold, s be more than or equal to the 1, natural number less than or equal to k,
First segment point set and second node collection can be any two nodes that user node collection, item nodes collection and label node are concentrated
Collection.
According to above formula (1) it is found that time for occurring of connection relation is closer to current time,It is smaller,It is bigger, corresponding weighted value wijAlso bigger；The number that connection relation occurs is more, and k is bigger, then corresponding
Weighted value wijAlso bigger.Weighted value is bigger, and bonding strength is bigger between node, shows that interest is stronger.
w0It can be true according to factors such as the production time of project, producer (for example, manufacturer, producer etc.), operators
Fixed, for example, for video, by taking the production time of project as an example, the newest video shown is to w0Value contribution it is bigger, otherwise then
It is smaller；By taking producer as an example, the video that the higher producer of popularity makes is to w0Otherwise value contribution is bigger then smaller.More than
Just for the sake of illustrating w0The method of determination of value, those skilled in the art can have according to actual application scenarios and demand
Body determines w0Value, the disclosure is not construed as limiting this.
According to above it will be appreciated that the initial connection set of relationship may include：Set AUI、AITAnd AUT.So according to weight
Value set W and initial connection set of relationship determine weighting connection relation set, can determine that weighting connection is closed according to following formula
Assembly is closed：
BUI=AUI×WUI, BIT=AIT×WIT, BUT=AUT×WUT。
Wherein, set BUIIndicate the weighting connection relation set between user node collection and item nodes collection, set BITTable
Weighting connection relation set between aspect mesh set of node and label node collection, set BUTIndicate user node collection and label node
Weighting connection relation set between collection.
For example, set BUIIn element dij=aij*w1ij, wherein w1ijFor set WUIIn element.Such as Fig. 2 institutes
Show,
Set BITIn element eij=bij*w2ij, wherein w2ijFor set WITIn element, set BUTIn element dij
=cij*w3ij, wherein w3ijFor set WUTIn element.Weighting connection relation set considers user interest migration, can be more preferable
Reflection user, project, the relationship of label three.
As shown in figure 4, according to embodiment of above it is found that step S12, according to initial resource and initial connection set of relations
It closes and carries out resource allocation acquisition resource allocation result, can specifically include：
Step S121 carries out resource allocation according to initial resource and weighting connection relation set and obtains resource allocation knot
Fruit.For example, according to initial resource and weighting connection relation set, broadcast algorithm can be based on or heat transfer algorithm carries out resource
Distribution obtains resource allocation result, is also based on heat transfer algorithm and carries out resource allocation acquisition resource allocation result, the disclosure
This is not construed as limiting.
The mode specifically distributed may refer to above example, repeat no more.
Fig. 5 shows the flow chart of the recommendation method according to one embodiment of the disclosure, as shown in figure 5, according to initial resource with
And weighting connection relation set carries out resource allocation and obtains resource allocation result, may include：
Step S1211 determines the first resource distribution of first direction according to initial resource and weighting connection relation set
As a result.
Step S1212 determines the Secondary resource distribution of second direction according to initial resource and weighting connection relation set
As a result.
For example, the first direction can be article-user-label-article direction, and the second direction can be
Article-label-user-article direction, alternatively, can also the first direction be article-label-user-article direction, it is described
Second direction is article-user-label-article direction.First direction and second direction can also be those skilled in the art's root
According to other directions for needing selection, the disclosure is without limitation.
The execution sequence of step S1211 and step S1212 are not construed as limiting, and be may be performed simultaneously, can also be first carried out wherein
One executes another again.Alternatively, one in step S1211 or step S1212 can also be only carried out, according only to initial resource
And weighting connection relation set determines that the resource allocation result in a direction, the disclosure are not construed as limiting this.
First direction can be determined based on broadcast algorithm, heat transfer algorithm etc. in the related technology, the resource of second direction is divided
With result, for example, by taking broadcast algorithm as an example, the first resource distribution of first direction can be determined according to following formula (2)
As a result：
Wherein, rsIt is for a certain user to the corresponding initial resources of project Is in the initial resource vector of allocation of items,
N, m, p correspond to the number of user node, item nodes and label node respectively.
The number of users of expression project Is adjoinings, i.e. article Is have connection side in user-project bipartite graph
Number of users, disIt is user-project connection relationship matrix BUIMiddle user UiElement corresponding with article Is,
Indicate that, in project-user direction, the resource of oneself is averagely given all users with connection relation therewith by each project,
It can obtain the corresponding resources of user node Ui.
Similarly,Indicate user UiThere is the number of labels on connection side, f in label-user's bipartite graphliIt is mark
Label-user's connection relationship matrix BUTMiddle user UiWith label TlCorresponding element,Indicate with
The resource of oneself is averagely given all labels marked, can obtain label node T by family-tag orientation, each userl
Corresponding resource.
Similarly,Indicate label TlThere is the number of articles on connection side, e in project-label bipartite graphjlIt is item
Mesh-label connection relationship matrix BITMiddle project Is and label TlCorresponding element,Expression is being marked
The resource of oneself is averagely given all projects with the label, can obtain project section by label-project direction, each label
Point IjCorresponding resource, j=1~m.
Each project after carrying out resource classification on first direction can be determined based on broadcast algorithm according to above formula (2)
The corresponding resource of node, and then first resource allocation result is obtained, the first resource allocation result can be to be calculated based on expansion
The corresponding resource vector R of item nodes collection after method distribution1。
The Secondary resource allocation result of second direction can be determined according to following formula (3)：
Detailed process may refer to the process of first direction distribution, repeat no more.
Each project after carrying out resource classification in second direction can be determined based on broadcast algorithm according to above formula (3)
The corresponding resource of node, and then Secondary resource allocation result is obtained, the Secondary resource allocation result can be to be calculated based on expansion
The corresponding resource vector R of item nodes collection after method distribution2。
Generating recommendation results according to resource allocation result can be, according to resource vector (R1Or R2) in corresponding element it is big
It is small to be ranked up, the project (can not also exclude) that user had been selected is excluded, final recommendation results are obtained, it is final to distribute
The resource arrived is more, illustrates that user's interest level is bigger, can be by corresponding project recommendation to user.
By taking shopping process as an example, with first direction：Article-user-label-article, which carries out resource allocation, can indicate to pass through
The corresponding article of label favored of user of purchase identical items is found to realize that article is recommended, such as, user 1 and user
2 all bought same electric cooker, and user 1 also has purchased certain hot-water bottle, the hot-water bottle that can be bought to 2 recommended user 1 of user；
And with second direction：Article-label-user-article, which carries out resource allocation, can indicate to favor article corresponding label by searching
The corresponding article of other labels favored of user realize that article is recommended, such as, user 1 and user 2 favor a certain
The corresponding label of article, user 1 also favor other labels, and the article corresponding to other labels that user 1 favors is recommended use
Family 2.Other directions are can also be, for example, carrying out resource allocation with user-article-label-user direction can indicate to pass through
Come to realize the recommendation of user by other users mark by the label corresponding to article that same subscriber was bought, for example, to
The other users crossed to the corresponding label for labelling of article that user 1 bought are recommended at family 1, and " user X likes identical object with you
Product add as a friend " etc..Therefore, the direction difference for carrying out resource allocation also just has different practical significances.
In a kind of possible embodiment, resource allocation is carried out according to initial resource and initial connection set of relationship and is obtained
Resource allocation result is obtained, can also include：
According to initial resource and initial connection set of relationship, the first resource allocation result of first direction is determined；
According to initial resource and initial connection set of relationship, the Secondary resource allocation result of second direction is determined.
Recommendation results can be generated according to first resource allocation result and Secondary resource allocation result.
For example, broadcast algorithm or heat transfer algorithm etc., which still can be based on, carries out resource allocation, specific resource allocation mistake
Journey may refer to described above, repeat no more.
As shown in figure 5, step S13, generates recommendation results according to resource allocation result, may include：
Step S131 generates recommendation results according to first resource classification results and Secondary resource allocation result.
For example, after the resource allocation in above-mentioned two direction, two resource vector R can be obtained1And R2(first
Resource classification result and Secondary resource allocation result), it is found that it can be according to the resource vector R of acquisition by above description1Or R2
(first resource classification results or Secondary resource allocation result) is directly to user's recommended project.In order to further increase the standard of recommendation
True property can also consider first resource classification results and Secondary resource allocation result to user's recommended project.
For example, recommendation results can be generated according to formula once (4)：
R*=λ × R1+(1-λ)×R2, (4)
Wherein λ ∈ [0,1], as λ=0, the money that is finally diffused according to article-label-user-article direction
Source vector R2As recommendation results, and as λ=1, finally it is diffused according to article-user-label-article direction
Resource vector R1As recommendation results.The value of λ can not be limited according to the selections such as the scene of application or actual demand, the disclosure
It is fixed.
Recommend method according to disclosure above-described embodiment, both direction resource point can be adjusted by adjusting the value of λ
It is carried with proportion according to recommending the demand of difference or recommendation etc. of method application scenarios to carry out more acurrate, effective recommendation
High user experience.
Fig. 6 shows the flow chart according to one exemplary recommendation method of the disclosure.As shown in fig. 6, it is possible, firstly, to data into
Row pretreatment, for example, removal is invalid, the data etc. of repetition；Secondly, user-project-label tripartite graph model of weighting is built,
It is specifically as follows：Complete tripartite graph G (V, E) is built based on user data, wherein V=(U, I, T), U are user node collection, I is
Item nodes integrate, T is label node collection, and E is that user node collection, item nodes collection and label node concentrate any two set of node
Between initial connection set of relationship；The time and number occurred according to connection relation determines the weighted value collection between each set of node
Close W；Weighting connection relation set is determined according to the weight value set and initial connection set of relationship；It is then possible to according to true
Fixed direction carries out initial resource, specially determines initial resource allocation value according to a certain node in a certain set of node；Then,
Resource re-allocation, which is carried out, according to the initial resource of distribution and weighting connection relation set obtains resource allocation result；Finally, root
It generates recommendation results according to resource allocation result and the recommendation results is sent to terminal device and show.
Embodiment 2
Fig. 7 shows that the block diagram of the recommendation apparatus according to one embodiment of the disclosure, the device can be applied to server etc..Such as
Shown in Fig. 7, which includes：Build module 71, resource distribution module 72 and recommendation results generation module 73.
Module 71 is built, for building complete tripartite graph G (V, E) based on user data, wherein V=(U, I, T), U are to use
Family set of node, I are that item nodes integrate, T is label node collection, and E is that user node collection, item nodes collection and label node concentration are appointed
Initial connection set of relationship between two sets of node of meaning；
Resource distribution module 72 is provided for carrying out resource allocation according to initial resource and initial connection set of relationship
Source allocation result；
Recommending module 73 generates recommendation results according to resource allocation result and the recommendation results is sent to terminal device
It is shown.
By introducing label, complete tripartite graph is built based on user data and remains initial connection between all sets of node
Relationship, and resource allocation is carried out according to the initial connection relationship and the initial resource of distribution and obtains resource allocation result, profit
It is generated recommendation results with the resource allocation result and the recommendation results is sent to terminal device and shown.According to this public affairs
The accuracy of recommendation can be improved by opening above-mentioned recommendation apparatus, realize personalized recommendation, improve user experience.
Fig. 8 shows the block diagram of the recommendation apparatus according to one embodiment of the disclosure, as shown in figure 8, the device further includes：Weight
Determining module 74 and weighting block 75.
Weight determination module 74, time and number for being occurred according to connection relation determine the weight between each set of node
Value set W；
Weighting block 75, for determining weighting connection relation according to the weight value set W and initial connection set of relationship
Set；
In a kind of possible embodiment, the recommending module 73 includes：First generation unit 731.
First generation unit 731, for carrying out resource allocation acquisition according to initial resource and weighting connection relation set
Resource allocation result.
In a kind of possible embodiment, the initial connection set of relationship includes：Set AUI, AITAnd AUT, described to add
Power connection relation, which combines, includes：Set BUI, BITAnd BUT,
The weighting block 75, including：Weighted units 751.
Weighted units 751, for determining weighting connection relation set according to following formula：
BUI=AUI×WUI, BIT=AIT×WIT, BUT=AUT×WUT,
Set AUIIndicate the initial connection set of relationship between user node collection and item nodes collection, set AITIndicate item
Initial connection set of relationship between mesh set of node and label node collection, set AUTIndicate user node collection and label node collection it
Between initial connection set of relationship；
WUIIndicate the weight value set between user node collection and item nodes collection, WITIndicate item nodes collection and label
Weight value set between set of node, WUTIndicate the weight value set between user node collection and label node collection.
In a kind of possible embodiment, the weight determination module 74 includes：Weight determining unit 741.
Weight determining unit 741, for determining the weight value set between each set of node according to following formula,
Wherein, wijFor the weighted value between i-th of node of first segment point set and j-th of node of second node collection, t is
Current time, k are the number that connection relation occurs, tij,sIt is j-th of i-th node and the second node collection of first segment point set
The time that each connection relation of node occurs, w0Indicate weight threshold, s is more than or equal to the 1, natural number less than or equal to k the
One set of node and second node collection are arbitrary two that the user node collection, the item nodes collection and the label node are concentrated
It is a.
In a kind of possible embodiment, the resource distribution module 72 includes：First allocation unit 721.
First allocation unit 721, for according to initial resource and weighting connection relation set, being based on broadcast algorithm or heat
It conducts algorithm and carries out resource allocation acquisition resource allocation result.
In a kind of possible embodiment, the resource distribution module 72 includes：First determination unit 722 and second is true
Order member 723.
First determination unit 722, for according to initial resource and weighting connection relation set, determining the of first direction
One resource allocation result；
Second determination unit 723, for according to initial resource and weighting connection relation set, determining the of second direction
Two resource allocation results；
The recommending module 73 includes：Second generation unit 732.
Second generation unit 732 recommends knot for being generated according to first resource allocation result and Secondary resource allocation result
Fruit.
In a kind of possible embodiment, the resource distribution module 72 includes：Third determination unit 724 and the 4th is really
Order member 725.
Third determination unit 724, for according to initial resource and initial connection set of relationship, determining first direction
One resource allocation result；
4th determination unit 725, for according to initial resource and initial connection set of relationship, determining second direction
Two resource allocation results,
The recommending module 73 includes：Third generation unit 733.
Third generation unit 733 recommends knot for being generated according to first resource allocation result and Secondary resource allocation result
Fruit.
In a kind of possible embodiment, the first direction is：Article-user-label-article direction；Described
Two directions are：Article-label-user-article direction.
Embodiment 3
Fig. 9 is a kind of block diagram of recommendation apparatus 1900 shown according to an exemplary embodiment.For example, device 1900 can be with
It is provided as a server.With reference to Fig. 9, device 1900 includes processing component 1922, further comprises one or more processing
Device and memory resource represented by a memory 1932, can be by the instruction of the execution of processing component 1922, example for storing
Such as application program.The application program stored in memory 1932 may include it is one or more each correspond to one group
The module of instruction.In addition, processing component 1922 is configured as executing instruction, to execute the above method.
Device 1900 can also include that a power supply module 1926 be configured as the power management of executive device 1900, one
Wired or wireless network interface 1950 is configured as device 1900 being connected to network and input and output (I/O) interface
1958.Device 1900 can be operated based on the operating system for being stored in memory 1932, such as Windows ServerTM, Mac
OS XTM, UnixTM, LinuxTM, FreeBSDTM or similar.
In the exemplary embodiment, it includes the non-volatile computer readable storage medium storing program for executing instructed, example to additionally provide a kind of
Such as include the memory 1932 of computer program instructions, above computer program instruction can be by the processing component 1922 of device 1900
It executes to complete the above method.
The disclosure can be system, method and/or computer program product.Computer program product may include computer
Readable storage medium storing program for executing, containing for making processor realize the computer-readable program instructions of various aspects of the disclosure.
Computer readable storage medium can be can keep and store the instruction used by instruction execution equipment tangible
Equipment.Computer readable storage medium for example can be-- but be not limited to-- storage device electric, magnetic storage apparatus, optical storage
Equipment, electromagnetism storage device, semiconductor memory apparatus or above-mentioned any appropriate combination.Computer readable storage medium
More specific example (non exhaustive list) includes：Portable computer diskette, random access memory (RAM), read-only is deposited hard disk
It is reservoir (ROM), erasable programmable read only memory (EPROM or flash memory), static RAM (SRAM), portable
Compact disk read-only memory (CD-ROM), digital versatile disc (DVD), memory stick, floppy disk, mechanical coding equipment, for example thereon
It is stored with punch card or groove internal projection structure and the above-mentioned any appropriate combination of instruction.Calculating used herein above
Machine readable storage medium storing program for executing is not interpreted that instantaneous signal itself, the electromagnetic wave of such as radio wave or other Free propagations lead to
It crosses the electromagnetic wave (for example, the light pulse for passing through fiber optic cables) of waveguide or the propagation of other transmission mediums or is transmitted by electric wire
Electric signal.
Computer-readable program instructions as described herein can be downloaded to from computer readable storage medium it is each calculate/
Processing equipment, or outer computer or outer is downloaded to by network, such as internet, LAN, wide area network and/or wireless network
Portion's storage device.Network may include copper transmission cable, optical fiber transmission, wireless transmission, router, fire wall, interchanger, gateway
Computer and/or Edge Server.Adapter or network interface in each calculating/processing equipment are received from network to be counted
Calculation machine readable program instructions, and the computer-readable program instructions are forwarded, for the meter being stored in each calculating/processing equipment
In calculation machine readable storage medium storing program for executing.
For execute the disclosure operation computer program instructions can be assembly instruction, instruction set architecture (ISA) instruction,
Machine instruction, machine-dependent instructions, microcode, firmware instructions, condition setup data or with one or more programming languages
Arbitrarily combine the source code or object code write, the programming language include the programming language-of object-oriented such as
Smalltalk, C++ etc., and conventional procedural programming languages-such as " C " language or similar programming language.Computer
Readable program instructions can be executed fully, partly execute on the user computer, is only as one on the user computer
Vertical software package executes, part executes or on the remote computer completely in remote computer on the user computer for part
Or it is executed on server.In situations involving remote computers, remote computer can pass through network-packet of any kind
It includes LAN (LAN) or wide area network (WAN)-is connected to subscriber computer, or, it may be connected to outer computer (such as profit
It is connected by internet with ISP).In some embodiments, by using computer-readable program instructions
Status information carry out personalized customization electronic circuit, such as programmable logic circuit, field programmable gate array (FPGA) or can
Programmed logic array (PLA) (PLA), the electronic circuit can execute computer-readable program instructions, to realize each side of the disclosure
Face.
Referring herein to according to the flow chart of the method, apparatus (system) of the embodiment of the present disclosure and computer program product and/
Or block diagram describes various aspects of the disclosure.It should be appreciated that flowchart and or block diagram each box and flow chart and/
Or in block diagram each box combination, can be realized by computer-readable program instructions.
These computer-readable program instructions can be supplied to all-purpose computer, special purpose computer or other programmable datas
The processor of processing unit, to produce a kind of machine so that these instructions are passing through computer or other programmable datas
When the processor of processing unit executes, work(specified in one or more of implementation flow chart and/or block diagram box is produced
The device of energy/action.These computer-readable program instructions can also be stored in a computer-readable storage medium, these refer to
It enables so that computer, programmable data processing unit and/or other equipment work in a specific way, to be stored with instruction
Computer-readable medium includes then a manufacture comprising in one or more of implementation flow chart and/or block diagram box
The instruction of the various aspects of defined function action.
Computer-readable program instructions can also be loaded into computer, other programmable data processing units or other
In equipment so that series of operation steps are executed on computer, other programmable data processing units or miscellaneous equipment, with production
Raw computer implemented process, so that executed on computer, other programmable data processing units or miscellaneous equipment
Instruct function action specified in one or more of implementation flow chart and/or block diagram box.
Flow chart and block diagram in attached drawing show the system, method and computer journey of multiple embodiments according to the disclosure
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
One module of table, program segment or a part for instruction, the module, program segment or a part for instruction include one or more use
The executable instruction of the logic function as defined in realization.In some implementations as replacements, the function of being marked in box
It can occur in a different order than that indicated in the drawings.For example, two continuous boxes can essentially be held substantially in parallel
Row, they can also be executed in the opposite order sometimes, this is depended on the functions involved.It is also noted that block diagram and/or
The combination of each box in flow chart and the box in block diagram and or flow chart can use function or dynamic as defined in executing
The dedicated hardware based system made is realized, or can be realized using a combination of dedicated hardware and computer instructions.
The presently disclosed embodiments is described above, above description is exemplary, and non-exclusive, and
It is not limited to disclosed each embodiment.Without departing from the scope and spirit of illustrated each embodiment, for this skill
Many modifications and changes will be apparent from for the those of ordinary skill in art field.The selection of term used herein, purport
In the principle, practical application or technological improvement to the technology in market for best explaining each embodiment, or this technology is made to lead
Other those of ordinary skill in domain can understand each embodiment disclosed herein.
Claims (18)
1. a kind of recommendation method, which is characterized in that including：
Complete tripartite graph G (V, E) is built based on user data, wherein V=(U, I, T), U are that user node integrates, I is project section
Point set, T are label node collection, and E is between user node collection, item nodes collection and label node concentration any two set of node
Initial connection set of relationship；
Resource allocation, which is carried out, according to initial resource and initial connection set of relationship obtains resource allocation result；
It generates recommendation results according to resource allocation result and the recommendation results is sent to terminal device and show.
2. recommendation method according to claim 1, which is characterized in that the method further includes：
The time and number occurred according to connection relation determines the weight value set W between each set of node；
Weighting connection relation set is determined according to the weight value set W and initial connection set of relationship；
Resource allocation, which is carried out, according to initial resource and initial connection set of relationship obtains resource allocation result, including：
Resource allocation, which is carried out, according to initial resource and weighting connection relation set obtains resource allocation result.
3. recommendation method according to claim 2, which is characterized in that the initial connection set of relationship includes：Set AUI,
AITAnd AUT, the weighting connection relation, which combines, includes：Set BUI, BITAnd BUT,
Weighting connection relation set is determined according to the weight value set W and initial connection set of relationship, including according to following public affairs
Formula determines weighting connection relation set：
BUI=AUI×WUI, BIT=AIT×WIT, BUT=AUT×WUT,
Set AUIIndicate the initial connection set of relationship between user node collection and item nodes collection, set AITIndicate item nodes
Initial connection set of relationship between collection and label node collection, set AUTIndicate first between user node collection and label node collection
Beginning connection relation set；
WUIIndicate the weight value set between user node collection and item nodes collection, WITIndicate item nodes collection and label node collection
Between weight value set, WUTIndicate the weight value set between user node collection and label node collection.
4. recommendation method according to claim 2 or 3, which is characterized in that the time occurred according to connection relation and number
Determine the weight value set W between each set of node, including：
The weight value set between each set of node is determined according to following formula,
Wherein, wijFor the weighted value between i-th of node of first segment point set and j-th of node of second node collection, t is current
Time, k are the number that connection relation occurs, tij,sFor j-th of node of i-th node and second node collection of first segment point set
The s times connection relation occur time, w0Indicate that weight threshold, s are more than or equal to the 1, natural number less than or equal to k first
Set of node and second node collection are arbitrary two that the user node collection, the item nodes collection and the label node are concentrated
It is a.
5. recommending method according to claim 2-4 any one of them, which is characterized in that connected according to initial resource and weighting
Set of relationship carries out resource allocation and obtains resource allocation result, including：
According to initial resource and weighting connection relation set, resource allocation acquisition is carried out based on broadcast algorithm or heat transfer algorithm
Resource allocation result.
6. recommending method according to claim 2-4 any one of them, which is characterized in that connected according to initial resource and weighting
Set of relationship carries out resource allocation and obtains resource allocation result, including：
According to initial resource and weighting connection relation set, the first resource allocation result of first direction is determined；
According to initial resource and weighting connection relation set, the Secondary resource allocation result of second direction is determined；
Recommendation results are generated according to resource allocation result, including：
Recommendation results are generated according to first resource allocation result and Secondary resource allocation result.
7. recommendation method according to claim 1, which is characterized in that according to initial resource and initial connection set of relationship
It carries out resource allocation and obtains resource allocation result, including：
According to initial resource and initial connection set of relationship, the first resource allocation result of first direction is determined；
According to initial resource and initial connection set of relationship, the Secondary resource allocation result of second direction is determined,
Recommendation results are generated according to resource allocation result, including：
Recommendation results are generated according to first resource allocation result and Secondary resource allocation result.
8. the recommendation method described according to claim 6 or 7, which is characterized in that
The first direction is：Article-user-label-article direction；
The second direction is：Article-label-user-article direction.
9. a kind of recommendation apparatus, which is characterized in that including：
Module is built, for building complete tripartite graph G (V, E) based on user data, wherein V=(U, I, T), U are user node
Integrate, I integrates as item nodes, T is label node collection, E is that user node collection, item nodes collection and label node concentrate any two
Initial connection set of relationship between set of node；
Resource distribution module obtains resource allocation for carrying out resource allocation according to initial resource and initial connection set of relationship
As a result；
Recommending module generates recommendation results according to resource allocation result and the recommendation results are sent to terminal device show
Show.
10. recommendation apparatus according to claim 9, which is characterized in that described device further includes：
Weight determination module, time and number for being occurred according to connection relation determine the weight value set between each set of node
W；
Weighting block, for determining weighting connection relation set according to the weight value set W and initial connection set of relationship；
The recommending module includes：
First generation unit obtains resource allocation for carrying out resource allocation according to initial resource and weighting connection relation set
As a result.
11. recommendation apparatus according to claim 10, which is characterized in that the initial connection set of relationship includes：Set
AUI, AITAnd AUT, the weighting connection relation, which combines, includes：Set BUI, BITAnd BUT,
The weighting block, including：
Weighted units, for determining weighting connection relation set according to following formula：
BUI=AUI×WUI, BIT=AIT×WIT, BUT=AUT×WUT,
Set AUIIndicate the initial connection set of relationship between user node collection and item nodes collection, set AITIndicate item nodes
Initial connection set of relationship between collection and label node collection, set AUTIndicate first between user node collection and label node collection
Beginning connection relation set；
WUIIndicate the weight value set between user node collection and item nodes collection, WITIndicate item nodes collection and label node collection
Between weight value set, WUTIndicate the weight value set between user node collection and label node collection.
12. the recommendation apparatus according to claim 10 or 11, which is characterized in that the weight determination module includes：
Weight determining unit, for determining the weight value set between each set of node according to following formula,
Wherein, wijFor the weighted value between i-th of node of first segment point set and j-th of node of second node collection, t is current
Time, k are the number that connection relation occurs, tij,sFor j-th of node of i-th node and second node collection of first segment point set
The s times connection relation occur time, w0Indicate that weight threshold, s are more than or equal to the 1, natural number less than or equal to k first
Set of node and second node collection are arbitrary two that the user node collection, the item nodes collection and the label node are concentrated
It is a.
13. according to claim 10-12 any one of them recommendation apparatus, which is characterized in that the resource distribution module includes：
First allocation unit, for according to initial resource and weighting connection relation set, being calculated based on broadcast algorithm or heat transfer
Method carries out resource allocation and obtains resource allocation result.
14. according to claim 10-12 any one of them recommendation apparatus, which is characterized in that the resource distribution module includes：
First determination unit, for according to initial resource and weighting connection relation set, determining the first resource of first direction
Allocation result；
Second determination unit, for according to initial resource and weighting connection relation set, determining the Secondary resource of second direction
Allocation result；
The recommending module includes：
Second generation unit, for generating recommendation results according to first resource allocation result and Secondary resource allocation result.
15. recommendation apparatus according to claim 9, which is characterized in that the resource distribution module includes：
Third determination unit, for according to initial resource and initial connection set of relationship, determining the first resource of first direction
Allocation result；
4th determination unit, for according to initial resource and initial connection set of relationship, determining the Secondary resource of second direction
Allocation result,
The recommending module includes：
Third generation unit, for generating recommendation results according to first resource allocation result and Secondary resource allocation result.
16. the recommendation apparatus according to claims 14 or 15, which is characterized in that
The first direction is：Article-user-label-article direction；
The second direction is：Article-label-user-article direction.
17. a kind of recommendation apparatus, which is characterized in that including：
Processor；
Memory for storing processor-executable instruction；
Wherein, the processor is configured as：
Complete tripartite graph G (V, E) is built based on user data, wherein V=(U, I, T), U are that user node integrates, I is project section
Point set, T are label node collection, and E is between user node collection, item nodes collection and label node concentration any two set of node
Initial connection set of relationship；
Resource allocation, which is carried out, according to initial resource and initial connection set of relationship obtains resource allocation result；
It generates recommendation results according to resource allocation result and the recommendation results is sent to terminal device and show.
18. a kind of non-volatile computer readable storage medium storing program for executing, is stored thereon with computer program instructions, which is characterized in that institute
State the method realized when computer program instructions are executed by processor described in any one of claim 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710300852.6A CN108805642A (en) | 2017-05-02 | 2017-05-02 | Recommend method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710300852.6A CN108805642A (en) | 2017-05-02 | 2017-05-02 | Recommend method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108805642A true CN108805642A (en) | 2018-11-13 |
Family
ID=64054286
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710300852.6A Pending CN108805642A (en) | 2017-05-02 | 2017-05-02 | Recommend method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108805642A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110264364A (en) * | 2019-04-30 | 2019-09-20 | 电子科技大学 | A kind of recommended method of investor |
| CN110750598A (en) * | 2019-09-18 | 2020-02-04 | 精锐视觉智能科技（深圳）有限公司 | Method and device for predicting article label, terminal equipment and storage medium |
| CN111143667A (en) * | 2019-12-05 | 2020-05-12 | 重庆邮电大学 | Three-part graph news recommendation method based on LDA topic model |
| CN111291904A (en) * | 2020-02-06 | 2020-06-16 | 腾讯科技（深圳）有限公司 | Preference prediction method and device and computer equipment |
| CN111400483A (en) * | 2020-03-17 | 2020-07-10 | 重庆邮电大学 | Time-weighting-based three-part graph news recommendation method |
| CN113743974A (en) * | 2021-01-14 | 2021-12-03 | 北京沃东天骏信息技术有限公司 | Resource recommendation method and device, equipment and storage medium |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080243785A1 (en) * | 2007-03-30 | 2008-10-02 | Tyron Jerrod Stading | System and methods of searching data sources |
| CN101853470A (en) * | 2010-05-28 | 2010-10-06 | 浙江大学 | Collaborative filtering method based on socialized label |
| CA2745536A1 (en) * | 2010-07-06 | 2012-01-06 | Omar M. Sheikh | Improving the relevancy of advertising material through user-defined preference filters, location and permission information |
| CN102376063A (en) * | 2011-11-29 | 2012-03-14 | 北京航空航天大学 | Social-label-based method for optimizing personalized recommendation system |
| CN104346476A (en) * | 2014-11-20 | 2015-02-11 | 西安电子科技大学 | Personalized object recommending method based on object similarity and network structure |
| CN104915861A (en) * | 2015-06-15 | 2015-09-16 | 浙江经贸职业技术学院 | An electronic commerce recommendation method for a user group model constructed based on scores and labels |
| CN106326367A (en) * | 2016-08-11 | 2017-01-11 | 华南师范大学 | Mixed collaborative recommendation algorithm based on WNBI and RSVD |
-
2017
- 2017-05-02 CN CN201710300852.6A patent/CN108805642A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080243785A1 (en) * | 2007-03-30 | 2008-10-02 | Tyron Jerrod Stading | System and methods of searching data sources |
| CN101853470A (en) * | 2010-05-28 | 2010-10-06 | 浙江大学 | Collaborative filtering method based on socialized label |
| CA2745536A1 (en) * | 2010-07-06 | 2012-01-06 | Omar M. Sheikh | Improving the relevancy of advertising material through user-defined preference filters, location and permission information |
| CN102376063A (en) * | 2011-11-29 | 2012-03-14 | 北京航空航天大学 | Social-label-based method for optimizing personalized recommendation system |
| CN104346476A (en) * | 2014-11-20 | 2015-02-11 | 西安电子科技大学 | Personalized object recommending method based on object similarity and network structure |
| CN104915861A (en) * | 2015-06-15 | 2015-09-16 | 浙江经贸职业技术学院 | An electronic commerce recommendation method for a user group model constructed based on scores and labels |
| CN106326367A (en) * | 2016-08-11 | 2017-01-11 | 华南师范大学 | Mixed collaborative recommendation algorithm based on WNBI and RSVD |
Non-Patent Citations (2)
| Title |
|---|
| ZI-KE ZHANG ET AL.: ""Personalized recommendation via integrated diffusion on user–item–tag tripartite graphs"", 《PHYSICA A》 * |
| 林举: ""基于完全三部图模型的个性化推荐技术研究"", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110264364A (en) * | 2019-04-30 | 2019-09-20 | 电子科技大学 | A kind of recommended method of investor |
| CN110264364B (en) * | 2019-04-30 | 2022-08-23 | 电子科技大学 | Recommendation method for investor |
| CN110750598A (en) * | 2019-09-18 | 2020-02-04 | 精锐视觉智能科技（深圳）有限公司 | Method and device for predicting article label, terminal equipment and storage medium |
| CN110750598B (en) * | 2019-09-18 | 2022-07-01 | 精锐视觉智能科技（深圳）有限公司 | Method and device for predicting article label, terminal equipment and storage medium |
| CN111143667A (en) * | 2019-12-05 | 2020-05-12 | 重庆邮电大学 | Three-part graph news recommendation method based on LDA topic model |
| CN111143667B (en) * | 2019-12-05 | 2023-05-02 | 重庆邮电大学 | Three-part graph news recommending method based on LDA topic model |
| CN111291904A (en) * | 2020-02-06 | 2020-06-16 | 腾讯科技（深圳）有限公司 | Preference prediction method and device and computer equipment |
| CN111291904B (en) * | 2020-02-06 | 2023-04-18 | 腾讯科技（深圳）有限公司 | Preference prediction method and device and computer equipment |
| CN111400483A (en) * | 2020-03-17 | 2020-07-10 | 重庆邮电大学 | Time-weighting-based three-part graph news recommendation method |
| CN111400483B (en) * | 2020-03-17 | 2022-06-21 | 重庆邮电大学 | Time-weighting-based three-part graph news recommendation method |
| CN113743974A (en) * | 2021-01-14 | 2021-12-03 | 北京沃东天骏信息技术有限公司 | Resource recommendation method and device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108805642A (en) | Recommend method and device | |
| Ngai et al. | Application of data mining techniques in customer relationship management: A literature review and classification | |
| CN111784455A (en) | Article recommendation method and recommendation equipment | |
| CN103902538B (en) | Information recommending apparatus and method based on decision tree | |
| Paul et al. | A quantitative and simulation model for managing sudden supply delay with fuzzy demand and safety stock | |
| CN106326318B (en) | Searching method and device | |
| CN109636430A (en) | Object identifying method and its system | |
| CN102385727A (en) | ID-value assessment device, ID-value assessment system, and ID-value assessment method | |
| CN113781149B (en) | Information recommendation method and device, computer readable storage medium and electronic equipment | |
| CN112989169B (en) | Target object identification method, information recommendation method, device, equipment and medium | |
| CN105740415B (en) | Bidding friend recommendation system based on label position weight and self study | |
| CN109993583A (en) | Information-pushing method and device, storage medium and electronic device | |
| Guchhait et al. | Inventory model of a deteriorating item with price and credit linked fuzzy demand: A fuzzy differential equation approach | |
| Krom | Global online entrepreneurship and the impact of innovation on brands | |
| Haque | Retail Demand Forecasting Using Neural Networks and Macroeconomic Variables | |
| CN112950321A (en) | Article recommendation method and device | |
| CN116362790A (en) | Client type prediction method, client type prediction device, electronic device, medium and program product | |
| Zhao et al. | Anatomy of a web-scale resale market: a data mining approach | |
| Pajo et al. | Fast lead user identification framework | |
| CN116342228B (en) | Related recommendation method based on directed graph neural network | |
| Hendalianpour et al. | Applying decision tree models to SMEs: A statistics-based model for customer relationship management | |
| Bhargavi et al. | Comparative study of consumer purchasing and decision pattern analysis using pincer search based data mining method | |
| CN110766488B (en) | Method and device for automatically determining theme scenes | |
| CN110807646A (en) | Data analysis method, device and computer readable storage medium | |
| CN111178974B (en) | Method and device for improving multi-platform fusion |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 100080 Beijing Haidian District city Haidian street A Sinosteel International Plaza No. 8 block 5 layer D Applicant after: YOUKU INFORMATION TECHNOLOGY (BEIJING) Co.,Ltd. Address before: 100080 Beijing Haidian District city Haidian street A Sinosteel International Plaza No. 8 block 5 layer D Applicant before: HEYI INFORMATION TECHNOLOGY (BEIJING) Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20200617 Address after: 310052 room 508, floor 5, building 4, No. 699, Wangshang Road, Changhe street, Binjiang District, Hangzhou City, Zhejiang Province Applicant after: Alibaba (China) Co.,Ltd. Address before: 100080 Beijing Haidian District city Haidian street A Sinosteel International Plaza No. 8 block 5 layer D Applicant before: YOUKU INFORMATION TECHNOLOGY (BEIJING) Co.,Ltd. |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181113 |