CN110209946A - Based on social and community Products Show method, system and storage medium - Google Patents
Based on social and community Products Show method, system and storage medium Download PDFInfo
- Publication number
- CN110209946A CN110209946A CN201910496478.0A CN201910496478A CN110209946A CN 110209946 A CN110209946 A CN 110209946A CN 201910496478 A CN201910496478 A CN 201910496478A CN 110209946 A CN110209946 A CN 110209946A
- Authority
- CN
- China
- Prior art keywords
- user
- product
- feedback
- preference
- community
- 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
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000003860 storage Methods 0.000 title claims abstract description 18
- 230000002452 interceptive effect Effects 0.000 claims abstract description 35
- 230000003993 interaction Effects 0.000 claims abstract description 26
- 238000012545 processing Methods 0.000 claims abstract description 8
- 230000000694 effects Effects 0.000 claims abstract description 6
- 125000004122 cyclic group Chemical group 0.000 claims description 6
- 238000000354 decomposition reaction Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 6
- 238000009826 distribution Methods 0.000 claims description 3
- PDEDQSAFHNADLV-UHFFFAOYSA-M potassium;disodium;dinitrate;nitrite Chemical compound [Na+].[Na+].[K+].[O-]N=O.[O-][N+]([O-])=O.[O-][N+]([O-])=O PDEDQSAFHNADLV-UHFFFAOYSA-M 0.000 claims description 3
- 239000000047 product Substances 0.000 abstract 7
- 239000006227 byproduct Substances 0.000 abstract 1
- 238000012549 training Methods 0.000 description 5
- 238000012360 testing method Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000015572 biosynthetic process Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 206010034719 Personality change Diseases 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- 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
-
- 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
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Economics (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Human Resources & Organizations (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Primary Health Care (AREA)
- Tourism & Hospitality (AREA)
- Development Economics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention provides a kind of Products Show method, system and storage medium based on social activity and community, is related to data processing field.The following steps are included: obtaining interaction data, social networks data and the community relation data of user and product, and history of forming data；Based on the historical data and pre-set partial ordering relation, the feedback relationship of user and product is determined；Determine user to the preference relation of product based on the historical data and the feedback relationship；Determine the objective function of the preference relation；User is obtained to the preference-score of product based on the objective function；It is ranked up based on non-interactive product of the preference-score to user, obtains Products Show result.The present invention can be accurately by Products Show to user.
Description
Technical field
The present invention relates to data processing fields, and in particular to a kind of based on social and community Products Show method, system
And storage medium.
Background technique
With the development of internet technology, the research of the personalized recommendation based on social networks is more active.And social activity pushes away
The method of recommending can be divided into the score in predicting algorithm based on explicit feedback data and the personalized ordering algorithm based on implicit feedback data.
Wherein, explicit feedback data refers generally to score, i.e. the scoring of 1-5, can significantly reflect user to the preference of product.And
Implicit feedback data refer to the interactive relation of user and product, such as browsing, purchase, collection, cannot significantly reflect user to production
The preference of product.For the product that those users and product do not interact, the product is not liked without it can be shown that user, it may
It is that user has not found the product.Since implicit feedback data are widely present, procurement cost is low, closer to reality, it is based on
The sort algorithm of implicit feedback data has obtained more and more concerns.
The social recommendation method that the prior art provides is based primarily upon Bayes's personalized ordering algorithm and carrys out Optimal scheduling.Pass through
The product interacted with user is set as positive feedback, the product that user did not interacted is set as negative sense feedback, and assumes user
The product not interacted is greater than to the preference of the product interacted, it is contemplated that good friend's social networks of user utilize Bayes's individual character
Change sort algorithm and then Products Show is optimized.
However the inventor of the present application discovered that above-mentioned recommended method is in practical application, user in online social networks
Social networks are relatively sparse, therefore only consider that social networks can not be accurately obtained the recommendation results of product.Therefore existing skill
The recommended method of art is not accurate enough.
Summary of the invention
(1) the technical issues of solving
In view of the deficiencies of the prior art, the present invention provides a kind of based on social and community Products Show method, system
And storage medium, solve the prior art can not accurate recommended products the technical issues of.
(2) technical solution
In order to achieve the above object, the present invention is achieved by the following technical programs:
The present invention solves a kind of based on social and community Products Show method, the recommendation provided by its technical problem
Method is executed by computer, comprising the following steps:
Obtain interaction data, social networks data and the community relation data of user and product, and history of forming data；
Based on the historical data and pre-set partial ordering relation, the feedback relationship of user and product is determined；
Determine user to the preference relation of product based on the historical data and the feedback relationship；
Determine the objective function of the preference relation；
User is obtained to the preference-score of product based on the objective function；
It is ranked up based on non-interactive product of the preference-score to user, obtains Products Show result.
Preferably, the historical data includes:
User set U, product set P, user and product had interactive set D, social networks set S and community relation
Set G.
Preferably, the pre-set partial ordering relation are as follows:
Positive feedback > social activity feedback > community feedback > negative-feedback；
The feedback relationship of the user and product includes:
Positive feedback set: PFu={ < u, pi> },
Wherein: < u, pi> ∈ D indicates user u and product piHad interactive；
Social feedback set: SFu={ < u, ps> },
Wherein: psIt is interactive to indicate that the social good friend of user u had, and the product that user u itself is not interacted；
Community feedback set: GFu={ < u, pg> },
Wherein: pgIt is interactive to indicate that the community good friend of user u had, and user u itself is good without the social activity of interaction and user u
The product that friend does not also interact；
Negative-feedback set: NFu={ < u, pj> },
Wherein: pjIndicate that user u itself is not interacted, the production that the social good friend and community good friend of user u does not also interact
Product.
Preferably, preference relation of the user to product are as follows:
Wherein:
Indicate user's set U to the sets of preferences of product set P；
The eigenmatrix of W expression user's set U；
The eigenmatrix of H expression product set P；
The bias term of b expression product set P.
Preferably, the objective function of the preference relation are as follows:
Wherein:
Indicate user u to positive feedback set PFuMiddle product piPreference；
Indicate user u to social feedback set SFuMiddle product psPreference；
Indicate user u to community feedback set GFuMiddle product pgPreference；
Indicate user u to negative-feedback set NFuMiddle product pjPreference；
σ () indicates logistic function；
Θ indicates the parameter sets in the matrix decomposition model, i.e. Θ={ W, H, b }；
λΘFor regularization parameter；
CSIndicate user u and product psIt does not interact, but had the good friend interacted in the good friend of user u with the product
Number；
CgIndicate user u and product pgNo interactions, and its social good friend is to the product no interactions, but the community of user u is good
There is interactive community good friend's number in friend to the product.
Preferably, user is obtained to the preference-score of product based on the objective function, specifically:
S501, the number of iterations is set as iter, wherein iter=1；
S502, normal distribution random initializtion parameter sets are based on:
Θiter={ Witer,Hiter,biter}
S503, cyclic variable is set as η, wherein η=1；
S504, under i-th ter times iteration traverse user and product interactive relation set D；
S505, in the η times cyclic process, randomly select a user u, while from the corresponding positive feedback collection of the user u
It closes, take out corresponding positive feedback product p in social feedback set, community feedback set and negative-feedback seti, social feedback product
ps, community feedback product pgWith negative-feedback product pj；To obtain one group of user's partial order of the η times traversal under i-th ter times iteration
Product mix
S506, by user's partial order product mixIt substitutes into the objective function, obtains i-th ter times repeatedly
The objective function recycled for lower the η
S507, objective function is updated based on stochastic gradient climb procedureMiddle parameterWithLadder
Degree；
S508, η+1 is assigned to η, and judges η > | D | it is whether true, if so, then follow the steps S509；Otherwise, it returns
Return step S505；
S509, judge parameterWhether it is convergence, if being to restrain, obtains optimized parameter collection
It closesOtherwise, iter+1 is assigned to iter, and return step S503.
Preferably, it is described obtain Products Show as a result, specifically:
Randomly select a user v in the product set U, according to the preference-score to user v in product set P
All non-interactive products carry out descending sorts, obtain Products Show result.
Preferably, further includes:
The Products Show result is assessed, the assessment includes: accuracy rate Precision@K assessment, recall rate
Recall@K assessment, MRR@K assessment, MAP@K assessment.
The present invention solves a kind of based on social and community Products Show system, the recommendation provided by its technical problem
System includes computer, and the computer includes:
At least one storage unit；
At least one processing unit；
Wherein, at least one instruction is stored at least one described storage unit, at least one instruction is by described
At least one processing unit is loaded and is executed to perform the steps of
Obtain interaction data, social networks data and the community relation data of user and product, and history of forming data；
Based on the historical data and pre-set partial ordering relation, the feedback relationship of user and product is determined；
Determine user to the preference relation of product based on the historical data and the feedback relationship；
Determine the objective function of the preference relation；
User is obtained to the preference-score of product based on the objective function；
It is ranked up based on non-interactive product of the preference-score to user, obtains Products Show result.
The present invention solves a kind of computer readable storage medium provided by its technical problem, is stored at least on the medium
One instruction, at least described instruction are loaded by processor and are executed to realize such as above-mentioned method
(3) beneficial effect
The present invention provides a kind of based on social and community Products Show method, system and storage medium.With existing skill
Art is compared, have it is following the utility model has the advantages that
The present invention is gone through by obtaining interaction data, social networks data and community relation data, the formation of user and product
History data；Based on historical data and pre-set partial ordering relation, the feedback relationship of user and product is determined；Based on historical data
Determine user to the preference relation of product with feedback relationship；And determine the objective function of preference relation；It is obtained based on objective function
Preference-score of the user to product；It is ranked up based on non-interactive product of the preference-score to user, obtains Products Show result.
The social networks and community relation of present invention combination user analyze the interaction of user and product, carry out to user preference
More fine-grained division, to keep recommendation results more accurate.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is the overall flow figure based on social and community Products Show method in the embodiment of the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, to the technology in the embodiment of the present invention
Scheme is clearly and completely described, it is clear that and described embodiments are some of the embodiments of the present invention, rather than whole
Embodiment.Based on the embodiments of the present invention, those of ordinary skill in the art are obtained without creative efforts
The every other embodiment obtained, shall fall within the protection scope of the present invention.
The embodiment of the present application is a kind of based on social and community Products Show method, system and storage medium by providing,
Solve the problems, such as the prior art can not accurate recommended products, realize the accurate recommendation of product.
Technical solution in the embodiment of the present application is in order to solve the above technical problems, general thought is as follows:
The embodiment of the present invention passes through interaction data, social networks data and the community relation data for obtaining user and product,
History of forming data；Based on historical data and pre-set partial ordering relation, the feedback relationship of user and product is determined；Based on going through
History data and feedback relationship determine user to the preference relation of product；And determine the objective function of preference relation；Based on target letter
Number obtains user to the preference-score of product；It is ranked up based on non-interactive product of the preference-score to user, obtains product and push away
Recommend result.The social networks and community relation of combination user of the embodiment of the present invention analyze the interaction of user and product, right
User preference has carried out more fine-grained division, to keep recommendation results more accurate.
In order to better understand the above technical scheme, in conjunction with appended figures and specific embodiments to upper
Technical solution is stated to be described in detail.
The embodiment of the invention provides a kind of based on social and community Products Show method, as shown in Figure 1, the above method
It is executed by computer, comprising the following steps:
S1, interaction data, social networks data and the community relation data for obtaining user and product, and history of forming number
According to；
S2, it is based on the historical data and pre-set partial ordering relation, determines the feedback relationship of user and product；
S3, determine user to the preference relation of product based on the historical data and the feedback relationship；
S4, the objective function for determining the preference relation；
S5, user is obtained to the preference-score of product based on the objective function；
S6, it is ranked up based on non-interactive product of the preference-score to user, obtains Products Show result.
The embodiment of the present invention passes through interaction data, social networks data and the community relation data for obtaining user and product,
History of forming data；Based on historical data and pre-set partial ordering relation, the feedback relationship of user and product is determined；Based on going through
History data and feedback relationship determine user to the preference relation of product；And determine the objective function of preference relation；Based on target letter
Number obtains user to the preference-score of product；It is ranked up based on non-interactive product of the preference-score to user, obtains product and push away
Recommend result.The social networks and community relation of combination user of the embodiment of the present invention analyze the interaction of user and product, right
User preference has carried out more fine-grained division, to keep recommendation results more accurate.
Each step is described in detail below:
In step sl, interaction data, social networks data and the community relation data of user and product are obtained, and are formed
Historical data.
Specifically, the embodiment of the present invention, which is based on internet platform, can obtain following data information:
User set U, product set P, user and product had interactive set D, social networks set S and community relation
Set G.
Set binary group < u, pi> ∈ D indicates user u and product piHad interactive.Wherein, u ∈ U, pi∈ P is shared and is used
Family | U |=M, product | P |=N number of.
Binary group < ui,uj> ∈ S indicates user uiWith user ujIt is good friend；Binary group < ui,uj> ∈ G indicates user ui
With user ujIt joined the same community.
In step s 2, it is based on the historical data and pre-set partial ordering relation, determines the feedback of user and product
Relationship.
Specifically, feedback relationship positive feedback set, social feedback set, community feedback set and negative-feedback set are come table
Show.
In practical applications, social networks being bi-directionally connected between good friend, belongs to strong ties.And community relation is then to use
Family is added according to the interest of oneself, belongs to unidirectional connection.Multiple community can be added in one user, and a community has multiple use
Family.And there are multiple interest by user itself, so the connection of the user belonged in same community between any two belongs to Weak link.Cause
This embodiment of the present invention presets feedback set with following partial ordering relation:
Positive feedback > social activity feedback > community feedback > negative-feedback.
Based on above-mentioned setting, feedback relationship can be obtained are as follows:
Positive feedback collection is combined into PFu={ < u, pi> }.
Wherein: < u, pi> ∈ D.
Social feedback set is SFu={ < u, ps> }.
Wherein: psIt is that the social good friend of user u had interactive, and the product that user u itself is not interacted.
Community feedback collection is combined into GFu={ < u, pg> }.
Wherein: pgIt is that the community good friend of user u had interactive, and user u itself is without interaction and the social good friend of user u
Also the product not interacted.
Negative-feedback collection is combined into NFu={ < u, pj> }.
Wherein: pjIt is the product that user u itself is not also interacted without the social and community good friend of interaction and user u.
Meet following conditions simultaneously with upper set:
PFu∪SFu∪GFu∪NFu=P
I.e. four feedback sets include all product data, and there do not have product data to exist simultaneously to be anti-in any two
In feedback set.
In step s3, determine user to the preference relation of product based on the historical data and the feedback relationship.
Specifically, indicating preference relation, preference relation tool based on matrix decomposition model of user's set U to product set P
Body are as follows:
Wherein:Indicate user's set U to the sets of preferences of product set P；
The eigenmatrix of W expression user's set U；
The eigenmatrix of H expression product set P；
The bias term of b expression product set P.
The dimension of W is M*K dimension, and the dimension of H is N*K dimension.
Based on known to the corresponding feedback set of above-mentioned preference relation:
Wherein:
Indicate user u to positive feedback set PFuMiddle product piPreference；
Indicate user u to social feedback set SFuMiddle product psPreference；
Indicate user u to community feedback set GFuMiddle product pgPreference；
Indicate user u to negative-feedback set NFuMiddle product pjPreference.
In step s 4, the objective function of the preference relation is determined.
Specifically, constructing objective function based on Bayes's personalized ordering method.It is closed matrix decomposition model as partial order
The measurement of system.It is optimized using gradient ascent algorithm, obtains the parameters value in matrix decomposition model.
Objective function specifically:
Wherein:
Indicate user u to positive feedback set PFuMiddle product piPreference；
Indicate user u to social feedback set SFuMiddle product psPreference；
Indicate user u to community feedback set GFuMiddle product pgPreference；
Indicate user u to negative-feedback set NFuMiddle product pjPreference；
σ () indicates logistic function；
Θ indicates the parameter sets in the matrix decomposition model, i.e. Θ={ W, H, b }；
λΘFor regularization parameter；
CSIndicate user u and product psIt does not interact, but had the good friend interacted in the good friend of user u with the product
Number；
CgIndicate user u and product pgNo interactions, and its social good friend is to the product no interactions, but the community of user u is good
There is interactive community good friend's number in friend to the product.
Wherein, CSValue is bigger, indicates user u to product piWith product psPreference it is closer；CgIt is worth bigger, expression user
To product psWith product pgPreference difference it is closer.
The purpose of the objective function is the preference-score for training user to product, is allowed to meet proposed by the present invention inclined
Order relation.
In step s 5, user is obtained to the preference-score of product based on the objective function.
Specifically, the following steps are included:
S501, the number of iterations is set as iter, wherein iter=1；
S502, normal distribution random initializtion parameter sets are based on:
Θiter={ Witer,Hiter,biter}
S503, cyclic variable is set as η, wherein η=1；
S504, under i-th ter times iteration traverse user and product interactive relation set D, to partial ordering relation carry out | D |
Secondary random sampling；
S505, in the η times cyclic process, randomly select a user u, while from the corresponding positive feedback collection of the user u
Close PFuIn randomly select a positive feedback product pi, from the corresponding social feedback set SF of the user uuIn randomly select a society
Hand over feedback product ps, from the corresponding community feedback set GF of the user uuIn randomly select a community feedback product pg, from the use
NF in the corresponding negative-feedback set of family uuRandomly select a negative-feedback product pj；To obtain under i-th ter times iteration the η times time
The one group of user's partial order product mix gone through
S506, by user's partial order product mixIt substitutes into the objective function, obtains i-th ter times
The objective function of the η circulation under iteration
S507, objective function is updated based on stochastic gradient climb procedureMiddle parameterWithLadder
Degree；
S508, η+1 is assigned to η, and judges η > | D | it is whether true, if so, then follow the steps S509；Otherwise, it returns
Return step S505；
S509, judge parameterWhether it is convergence, if being to restrain, obtains optimized parameter collection
It closesOtherwise, iter+1 is assigned to iter, and return step S503.
In step s 6, it is ranked up based on non-interactive product of the preference-score to user, obtains Products Show knot
Fruit.
Specifically, a user v in the product set U is randomly selected, according to the preference-score to user v in product
All non-interactive products carry out descending sort in set P, can be preset to extract preceding n product formation Products Show column
Table obtains Products Show result and recommends user.
The embodiment of the present invention considers social information and community information, constructs use on the basis of user interacts with product
Family is to the preference ordering of non-interactive product, so that user preference has been carried out more fine-grained division.
The embodiment of the invention also includes:
S7, the Products Show result is assessed.
Specifically, the embodiment of the present invention is commented using accuracy rate Precision@K, recall rate Recall@K, MRR@K, MAP@K
Estimate recommendation results.
Firstly, data set is divided into training set and test set.The present invention uses training set training pattern, obtains user couple
The preference-score x of product.Obtain the recommendation list of each user according to preference-score, the recommendation list that is obtained using training set and
Test set compares assessment, to obtain the quality of recommendation results.
Wherein, ratio shared by the article that user likes in precision@K expression recommendation list.
For the recommendation accuracy rate of single user u are as follows:
Wherein:
It is user's u actual purchase that how many is a in the recommendation list of hit expression user u；
K is recommendation results length.
Overall accuracy rate are as follows:
I.e. the predictablity rate of population sample is being averaged for personal accuracy rate.
The article that Recall@K indicates that how many user likes in test set appears in recommendation list.
For the recommendation recall rate of single user u are as follows:
Wherein:
It is user's u actual purchase that how many is a in the recommendation list of hit expression user u.
For the number of the product of user's u actual purchase.
Overall recall rate are as follows:
I.e. the recall rate of population sample is being averaged for personal recall rate.
MRR@K specifically:
Wherein:
rankuIndicate the position in the recommendation list of user u where first accurate product of prediction.
MAP@K specifically:
Wherein:
Hit indicates how many a product for having hit user's u actual purchase in the recommendation list of user u.
The embodiment of the invention also provides a kind of based on social and community Products Show system, and the system comprises calculating
Machine, the computer include:
At least one storage unit；
At least one processing unit；
Wherein, at least one instruction is stored at least one described storage unit, at least one instruction is by described
At least one processing unit is loaded and is executed to perform the steps of
Obtain interaction data, social networks data and the community relation data of user and product, and history of forming data；
Based on the historical data and pre-set partial ordering relation, the feedback relationship of user and product is determined；
Determine user to the preference relation of product based on the historical data and the feedback relationship；
Determine the objective function of the preference relation；
User is obtained to the preference-score of product based on the objective function；
It is ranked up based on non-interactive product of the preference-score to user, obtains Products Show result.
It will be appreciated that above-mentioned recommender system provided in an embodiment of the present invention is corresponding with above-mentioned recommended method, it is related
It the part such as explanation, citing, beneficial effect of content can be with reference to corresponding interior in the Products Show method based on social and community
Hold, is not repeating herein.
The embodiment of the present invention also provides a kind of computer readable storage medium, is stored at least one in above-mentioned storage medium
Instruction, an at least the above instruction are loaded by processor and are executed to realize such as the above method.
In conclusion compared with prior art, have it is following the utility model has the advantages that
The embodiment of the present invention passes through interaction data, social networks data and the community relation data for obtaining user and product,
History of forming data；Based on historical data and pre-set partial ordering relation, the feedback relationship of user and product is determined；Based on going through
History data and feedback relationship determine user to the preference relation of product；And determine the objective function of preference relation；Based on target letter
Number obtains user to the preference-score of product；It is ranked up based on non-interactive product of the preference-score to user, obtains product and push away
Recommend result.The social networks and community relation of combination user of the embodiment of the present invention analyze the interaction of user and product, right
User preference has carried out more fine-grained division, to keep recommendation results more accurate.
It should be noted that through the above description of the embodiments, those skilled in the art can be understood that
It can be realized by means of software and necessary general hardware platform to each embodiment.Based on this understanding, above-mentioned skill
Substantially the part that contributes to existing technology can be embodied in the form of software products art scheme in other words, the calculating
Machine software product may be stored in a computer readable storage medium, such as ROM/RAM, magnetic disk, CD, including some instructions are used
So that computer equipment (can be personal computer, server or the network equipment etc.) execute each embodiment or
Method described in certain parts of person's embodiment.
In the instructions provided here, numerous specific details are set forth.It is to be appreciated, however, that implementation of the invention
Example can be practiced without these specific details.In some instances, well known method, structure is not been shown in detail
And technology, so as not to obscure the understanding of this specification.
Herein, relational terms such as first and second and the like be used merely to by an entity or operation with it is another
One entity or operation distinguish, and without necessarily requiring or implying between these entities or operation, there are any this reality
Relationship or sequence.Moreover, the terms "include", "comprise" or its any other variant are intended to the packet of nonexcludability
Contain, so that the process, method, article or equipment for including a series of elements not only includes those elements, but also including
Other elements that are not explicitly listed, or further include for elements inherent to such a process, method, article, or device.
In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including the element
Process, method, article or equipment in there is also other identical elements.
The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although with reference to the foregoing embodiments
Invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each implementation
Technical solution documented by example is modified or equivalent replacement of some of the technical features；And these modification or
Replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution.
Claims (10)
1. a kind of based on social and community Products Show method, which is characterized in that the recommended method is executed by computer, packet
Include following steps:
Obtain interaction data, social networks data and the community relation data of user and product, and history of forming data；
Based on the historical data and pre-set partial ordering relation, the feedback relationship of user and product is determined；
Determine user to the preference relation of product based on the historical data and the feedback relationship；
Determine the objective function of the preference relation；
User is obtained to the preference-score of product based on the objective function；
It is ranked up based on non-interactive product of the preference-score to user, obtains Products Show result.
2. recommended method as described in claim 1, which is characterized in that the historical data includes:
User set U, product set P, user and product had interactive set D, social networks set S and community relation set
G。
3. recommended method as claimed in claim 2, which is characterized in that the pre-set partial ordering relation are as follows:
Positive feedback > social activity feedback > community feedback > negative-feedback；
The feedback relationship of the user and product includes:
Positive feedback set: PFu={ < u, pi> },
Wherein: < u, pi> ∈ D indicates user u and product piHad interactive；
Social feedback set: SFu={ < u, ps> },
Wherein: psIt is interactive to indicate that the social good friend of user u had, and the product that user u itself is not interacted；
Community feedback set: GFu={ < u, pg> },
Wherein: pgIt is interactive to indicate that the community good friend of user u had, and user u itself is without interaction and the social good friend of user u
The product not interacted；
Negative-feedback set: NFu={ < u, pj> },
Wherein: pjIndicate that user u itself is not interacted, the product that the social good friend and community good friend of user u does not also interact.
4. recommended method as claimed in claim 3, which is characterized in that preference relation of the user to product are as follows:
Wherein:
Indicate user's set U to the sets of preferences of product set P；
The eigenmatrix of W expression user's set U；
The eigenmatrix of H expression product set P；
The bias term of b expression product set P.
5. recommended method as claimed in claim 4, which is characterized in that the objective function of the preference relation are as follows:
Wherein:
Indicate user u to positive feedback set PFuMiddle product piPreference；
Indicate user u to social feedback set SFuMiddle product psPreference；
Indicate user u to community feedback set GFuMiddle product pgPreference；
Indicate user u to negative-feedback set NFuMiddle product pjPreference；
σ () indicates logistic function；
Θ indicates the parameter sets in the matrix decomposition model, i.e. Θ={ W, H, b }；
λΘFor regularization parameter；
CSIndicate user u and product psIt does not interact, but had the good friend's number interacted in the good friend of user u with the product；
CgIndicate user u and product pgNo interactions, and its social good friend is to the product no interactions, but in the community good friend of user u
There is interactive community good friend's number to the product.
6. recommended method as claimed in claim 5, which is characterized in that obtain user to the inclined of product based on the objective function
Good score, specifically:
S501, the number of iterations is set as iter, wherein iter=1；
S502, normal distribution random initializtion parameter sets are based on:
Θiter={ Witer,Hiter,biter}
S503, cyclic variable is set as η, wherein η=1；
S504, under i-th ter times iteration traverse user and product interactive relation set D；
S505, in the η times cyclic process, randomly select a user u, while from the corresponding positive feedback set of the user u, society
It hands in feedback set, community feedback set and negative-feedback set and takes out corresponding positive feedback product pi, social feedback product ps, society
Group's feedback product pgWith negative-feedback product pj；To obtain one group of user's partial order product group of the η times traversal under i-th ter times iteration
It closes
S506, by user's partial order product mixIt substitutes into the objective function, obtains under i-th ter times iteration
The objective function of the η circulation
S507, objective function is updated based on stochastic gradient climb procedureMiddle parameterWithGradient；
S508, η+1 is assigned to η, and judges η > | D | it is whether true, if so, then follow the steps S509；Otherwise, step is returned
Rapid S505；
S509, judge parameterWhether it is convergence, if being to restrain, obtains optimized parameter setOtherwise, iter+1 is assigned to iter, and return step S503.
7. recommended method as claimed in claim 2, which is characterized in that it is described obtain Products Show as a result, specifically:
Randomly select a user v in the product set U, the institute according to the preference-score to user v in product set P
There is non-interactive product to carry out descending sort, obtains Products Show result.
8. recommended method as described in claim 1, which is characterized in that further include:
The Products Show result is assessed, the assessment includes: accuracy rate Precision@K assessment, recall rate
Recall@K assessment, MRR@K assessment, MAP@K assessment.
9. a kind of based on social and community Products Show system, which is characterized in that the recommender system includes computer, described
Computer includes:
At least one storage unit；
At least one processing unit；
Wherein, be stored at least one instruction at least one described storage unit, at least one instruction by it is described at least
One processing unit is loaded and is executed to perform the steps of
Obtain interaction data, social networks data and the community relation data of user and product, and history of forming data；
Based on the historical data and pre-set partial ordering relation, the feedback relationship of user and product is determined；
Determine user to the preference relation of product based on the historical data and the feedback relationship；
Determine the objective function of the preference relation；
User is obtained to the preference-score of product based on the objective function；
It is ranked up based on non-interactive product of the preference-score to user, obtains Products Show result.
10. a kind of computer readable storage medium, be stored at least one instruction on the medium, at least described instruction by
Reason device is loaded and is executed to realize the method as described in claim 1.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910496478.0A CN110209946B (en) | 2019-06-10 | 2019-06-10 | Social and community-based product recommendation method, system and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910496478.0A CN110209946B (en) | 2019-06-10 | 2019-06-10 | Social and community-based product recommendation method, system and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110209946A true CN110209946A (en) | 2019-09-06 |
| CN110209946B CN110209946B (en) | 2021-03-09 |
Family
ID=67791600
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910496478.0A Active CN110209946B (en) | 2019-06-10 | 2019-06-10 | Social and community-based product recommendation method, system and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110209946B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110837598A (en) * | 2019-11-11 | 2020-02-25 | 腾讯科技（深圳）有限公司 | Information recommendation method, device, equipment and storage medium |
| CN111667298A (en) * | 2020-04-15 | 2020-09-15 | 合肥工业大学 | User personalized demand prediction method and system |
| CN112149003A (en) * | 2020-10-28 | 2020-12-29 | 浙江集享电子商务有限公司 | Commodity community recommendation method and device and computer equipment |
| CN112231579A (en) * | 2019-12-30 | 2021-01-15 | 北京邮电大学 | Social video recommendation system and method based on implicit community discovery |
| CN116594712A (en) * | 2023-05-29 | 2023-08-15 | 上海炫稷网络科技有限公司 | User center display method and system based on data analysis |
| CN117312681A (en) * | 2023-08-23 | 2023-12-29 | 苏州大学 | Meta universe oriented user preference product recommendation method and system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120123837A1 (en) * | 2010-11-17 | 2012-05-17 | Wellstar Gmbh & Co Kg | Social network shopping system and method |
| WO2014070293A1 (en) * | 2012-11-05 | 2014-05-08 | Nara Logics, Inc. | Systems and methods for providing enhanced neural network genesis and recommendations to one or more users |
| CN106777123A (en) * | 2016-12-16 | 2017-05-31 | 合肥工业大学 | A kind of group based on two-way tensor resolution model recommends method |
| CN107391637A (en) * | 2017-07-10 | 2017-11-24 | 江苏省现代企业信息化应用支撑软件工程技术研发中心 | For possessing the group recommending method of geographical social information |
| CN107833117A (en) * | 2017-12-13 | 2018-03-23 | 合肥工业大学 | A kind of Bayes's personalized ordering for considering label information recommends method |
| CN108763314A (en) * | 2018-04-26 | 2018-11-06 | 深圳市腾讯计算机系统有限公司 | A kind of interest recommends method, apparatus, server and storage medium |
-
2019
- 2019-06-10 CN CN201910496478.0A patent/CN110209946B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120123837A1 (en) * | 2010-11-17 | 2012-05-17 | Wellstar Gmbh & Co Kg | Social network shopping system and method |
| WO2014070293A1 (en) * | 2012-11-05 | 2014-05-08 | Nara Logics, Inc. | Systems and methods for providing enhanced neural network genesis and recommendations to one or more users |
| CN106777123A (en) * | 2016-12-16 | 2017-05-31 | 合肥工业大学 | A kind of group based on two-way tensor resolution model recommends method |
| CN107391637A (en) * | 2017-07-10 | 2017-11-24 | 江苏省现代企业信息化应用支撑软件工程技术研发中心 | For possessing the group recommending method of geographical social information |
| CN107833117A (en) * | 2017-12-13 | 2018-03-23 | 合肥工业大学 | A kind of Bayes's personalized ordering for considering label information recommends method |
| CN108763314A (en) * | 2018-04-26 | 2018-11-06 | 深圳市腾讯计算机系统有限公司 | A kind of interest recommends method, apparatus, server and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| TONG ZHAO等: "Leveraging Social Connections to Improve Personalized Ranking for Collaborative Filtering", 《THE 23RD ACM INTERNATIONAL CONFERENCE》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110837598A (en) * | 2019-11-11 | 2020-02-25 | 腾讯科技（深圳）有限公司 | Information recommendation method, device, equipment and storage medium |
| CN112231579A (en) * | 2019-12-30 | 2021-01-15 | 北京邮电大学 | Social video recommendation system and method based on implicit community discovery |
| CN112231579B (en) * | 2019-12-30 | 2022-10-28 | 北京邮电大学 | Social video recommendation system and method based on implicit community discovery |
| CN111667298A (en) * | 2020-04-15 | 2020-09-15 | 合肥工业大学 | User personalized demand prediction method and system |
| CN111667298B (en) * | 2020-04-15 | 2022-09-30 | 合肥工业大学 | User personalized demand prediction method and system |
| CN112149003A (en) * | 2020-10-28 | 2020-12-29 | 浙江集享电子商务有限公司 | Commodity community recommendation method and device and computer equipment |
| CN116594712A (en) * | 2023-05-29 | 2023-08-15 | 上海炫稷网络科技有限公司 | User center display method and system based on data analysis |
| CN116594712B (en) * | 2023-05-29 | 2024-02-23 | 上海炫稷网络科技有限公司 | User center display method and system based on data analysis |
| CN117312681A (en) * | 2023-08-23 | 2023-12-29 | 苏州大学 | Meta universe oriented user preference product recommendation method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110209946B (en) | 2021-03-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110209946A (en) | Based on social and community Products Show method, system and storage medium | |
| CN104063481B (en) | A kind of film personalized recommendation method based on the real-time interest vector of user | |
| CN111797321B (en) | Personalized knowledge recommendation method and system for different scenes | |
| CN108431833A (en) | End-to-end depth collaborative filtering | |
| CN106682114A (en) | Personalized recommending method fused with user trust relationships and comment information | |
| CN104794367B (en) | Medical treatment resource scoring based on hidden semantic model is with recommending method | |
| CN107491813A (en) | A kind of long-tail group recommending method based on multiple-objection optimization | |
| CN104361062B (en) | A kind of recommendation method and device of related information | |
| CN105354260B (en) | The Mobile solution of a kind of mosaic society's network and item characteristic recommends method | |
| Bin et al. | Collaborative filtering recommendation algorithm based on multi-relationship social network | |
| CN105354330A (en) | Sparse data preprocessing based collaborative filtering recommendation method | |
| CN108038629A (en) | A kind of optimization method based on collaborative filtering | |
| CN104837184A (en) | Heterogeneous wireless network selection method based on interval triangular fuzzy number | |
| CN107679239A (en) | Recommend method in a kind of personalized community based on user behavior | |
| Do et al. | Dynamic weighted hybrid recommender systems | |
| CN106127506A (en) | A kind of recommendation method solving commodity cold start-up problem based on Active Learning | |
| CN110059257B (en) | Project recommendation method based on score correction | |
| CN109190040A (en) | Personalized recommendation method and device based on coevolution | |
| CN107909498B (en) | Recommendation method based on area below maximized receiver operation characteristic curve | |
| Cheng et al. | Collaborative filtering algorithm based on data mixing and filtering | |
| Insuwan et al. | Improving missing values imputation in collaborative filtering with user-preference genre and singular value decomposition | |
| CN107844585A (en) | A kind of recommendation method for considering overlapping community's effect | |
| CN108763515B (en) | Time-sensitive personalized recommendation method based on probability matrix decomposition | |
| CN106777123A (en) | A kind of group based on two-way tensor resolution model recommends method | |
| KR20200122652A (en) | Nutrient Profiling-based Pet Food Recommendation System |
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 |