CN110781389A - Method and system for generating recommendations for a user - Google Patents
Method and system for generating recommendations for a user Download PDFInfo
- Publication number
- CN110781389A CN110781389A CN201910994809.3A CN201910994809A CN110781389A CN 110781389 A CN110781389 A CN 110781389A CN 201910994809 A CN201910994809 A CN 201910994809A CN 110781389 A CN110781389 A CN 110781389A
- Authority
- CN
- China
- Prior art keywords
- social
- user
- view
- model
- recommendation
- 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
- 238000000034 method Methods 0.000 title claims abstract description 48
- 238000012549 training Methods 0.000 claims abstract description 8
- 230000006870 function Effects 0.000 claims description 51
- 230000006855 networking Effects 0.000 claims description 22
- 239000011159 matrix material Substances 0.000 claims description 16
- 238000012546 transfer Methods 0.000 claims description 13
- 238000000354 decomposition reaction Methods 0.000 claims description 8
- 238000011478 gradient descent method Methods 0.000 claims description 6
- 238000011176 pooling Methods 0.000 claims description 6
- 238000012935 Averaging Methods 0.000 claims description 5
- 230000010365 information processing Effects 0.000 claims description 2
- 239000013598 vector Substances 0.000 description 12
- 238000010586 diagram Methods 0.000 description 11
- 238000004891 communication Methods 0.000 description 7
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000014509 gene expression Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
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
- G06F16/9536—Search customisation based on social or collaborative filtering
-
- 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)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Marketing (AREA)
- General Engineering & Computer Science (AREA)
- Finance (AREA)
- Data Mining & Analysis (AREA)
- Accounting & Taxation (AREA)
- General Business, Economics & Management (AREA)
- Strategic Management (AREA)
- Primary Health Care (AREA)
- Computing Systems (AREA)
- Tourism & Hospitality (AREA)
- Human Resources & Organizations (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Development Economics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The present application relates to a method for generating recommendations for a user, the method comprising: generating a recommendation model, wherein the recommendation model is used for recommending articles for a user; generating a multi-view social model, the multi-view social model corresponding to a plurality of social relationship types; combining the recommendation model with the multi-view social model to arrive at a multi-view social recommendation model; training the multi-view social recommendation model; and generating a recommendation for the user using the trained multi-view social recommendation model. Related systems and computer storage media are also disclosed. The recommendation method and the recommendation device can better generate recommendations for the user.
Description
Technical Field
The present invention relates to recommendations, and more particularly, to a method and system for generating recommendations for a user.
Background
Currently, social networks have become increasingly popular. For example, specialized social networking sites such as WeChat, microblog, etc. provide social networking services. In addition, services such as Payment treasures, Office 365, etc. also support social networking functionality. In these social networks, social relationships exist between users.
With the popularity of social networks, social network-based recommendation systems have also been developed. For example, in some social networks, content is recommended to a user based on content viewed by friends of the user. For another example, in some social networks, items are recommended to a user based on items of interest to the user's relatives.
However, in current social network-based recommendation systems, only one type of social relationship is typically considered. In part because some social networks have only a single type of social relationship. For example, microblogs only have concerns. However, even in systems that own or have access to multiple types of network relationships of users, existing systems still only consider one type of social relationship. Considering only one social relationship makes the recommendations generated for the user insufficiently accurate.
Disclosure of Invention
In order to improve the accuracy of social network recommendation, the invention provides a method and a system for generating recommendations for a user.
The invention achieves the above purpose through the following technical scheme.
In one aspect, a method for generating recommendations for a user is disclosed, the method comprising:
generating a recommendation model, wherein the recommendation model is used for recommending articles for a user;
generating a multi-view social model, the multi-view social model corresponding to a plurality of social relationship types;
combining the recommendation model with the multi-view social model to arrive at a multi-view social recommendation model;
training the multi-view social recommendation model; and
generating recommendations for the user using a trained multi-view social recommendation model.
Preferably, generating the multi-view social model comprises:
generating a plurality of single-view social models, each single-view social model corresponding to a social relationship type;
combining the plurality of single-view social models to arrive at a multi-view social model.
Preferably, the plurality of single-view social models correspond to a plurality of social networking services.
Preferably, each of the plurality of single-view social models corresponds to a different social networking service.
Preferably, the plurality of social relationship types includes two or more of a friendship, an affinity, a correspondence, and a transfer relationship.
Preferably, combining the plurality of single-view social models comprises summing objective functions of the plurality of single-view social models.
Preferably, combining the plurality of single-view social models comprises averaging an objective function of the plurality of single-view social models.
Preferably, combining the plurality of single-view social models comprises weighted summation of objective functions of the plurality of single-view social models.
Preferably, combining the plurality of single-view social models comprises performing a pooling operation on objective functions of the plurality of single-view social models.
Preferably, combining the plurality of single-view social models comprises determining a combining function for combining the plurality of single-view social models.
Preferably, generating a recommendation for the user using the trained multi-view social recommendation model comprises:
acquiring user information of the user, wherein the user information comprises known scoring information of the user on an article and social information of a plurality of social relation types between the user and the user;
determining a score for a plurality of candidate items using a trained multi-view social recommendation model based on user information of the user; and
recommending items to the user based on the scores of the plurality of candidate items.
Preferably, the recommendation model is a matrix decomposition model.
Preferably, training the multi-view social recommendation model comprises: iteratively solving an objective function of the multi-view social recommendation model using a gradient descent method to determine parameters of the multi-view social recommendation model.
Preferably, generating a recommendation for the user using the trained multi-view social recommendation model comprises: a score for the candidate item is determined using the determined parameters.
Preferably, generating a recommendation for the user comprises generating a recommended good or recommended content for the user.
In another aspect, a system for generating recommendations for a user is disclosed, the system comprising:
the system comprises a user information acquisition module, a user information acquisition module and a user information processing module, wherein the user information acquisition module is used for acquiring user information of a user, and the user information comprises known scoring information of the user on an article and social information of a plurality of social relationship types between the user and the user;
a multi-view social recommendation module to determine scores for a plurality of candidate items based on user information of the user, the multi-view social recommendation module comprising a recommendation module and a multi-view social module, wherein:
the recommendation module implements a recommendation model for recommending items for a user,
the multi-view social recommendation module includes a plurality of single-view social modules that implement a plurality of single-view social models corresponding to a plurality of social relationship types, and a combining module to combine the plurality of single-view social modules into the multi-view social module.
In yet another aspect, a computer-readable storage medium storing instructions that, when executed by a computer, cause the computer to perform the method as described above is disclosed.
In yet another aspect, a system is disclosed that includes means for performing the method described above.
Compared with the prior art, the invention has the following beneficial effects:
the method and the device can make full use of various different types of social relations to generate recommendations for the user more accurately.
Drawings
The foregoing summary, as well as the following detailed description of the invention, will be better understood when read in conjunction with the appended drawings. It is to be noted that the appended drawings are intended as examples of the claimed invention. In the drawings, like reference characters designate the same or similar elements.
FIG. 1A is a diagram illustrating a social graph with a single view.
FIG. 1B is a diagram illustrating a social graph with multiple views.
FIG. 2 illustrates a schematic diagram of a single-view social recommendation model.
FIG. 3 illustrates a schematic diagram of a multi-view social recommendation model in accordance with an embodiment of the present description.
FIG. 4 is a flow diagram illustrating a method for generating recommendations for a user in accordance with an embodiment of the present invention.
FIG. 5 shows a schematic diagram of a system for generating recommendations for a user, according to an embodiment of the invention.
Detailed Description
The detailed features and advantages of the present invention are described in detail in the detailed description which follows, and will be sufficient for anyone skilled in the art to understand the technical content of the present invention and to implement the present invention, and the related objects and advantages of the present invention will be easily understood by those skilled in the art from the description, claims and drawings disclosed in the present specification.
In order to fully utilize various types of social relationships in a social networking system and improve the accuracy of recommendations, the invention provides a method, a system and a computer readable medium capable of generating recommendations for a user by a multi-view social model.
Social networking or social networking services referred to herein include, but are not limited to, purely social networking sites (e.g., WeChat, microblog, etc.), as well as any other social networking enabled system (e.g., Payment Bao, etc.).
Social relationships referred to in this application include, but are not limited to, friendships, relationships, transfer relationships, communication relationships, and the like.
Hereinafter, the basic concept of the social recommendation model is first introduced, then the recommendation model and the social model are introduced separately, then the single-view social recommendation model and the multi-view social recommendation model are introduced, and finally the multi-view social recommendation method according to the embodiment of the present invention is introduced.
Social recommendation model
The social recommendation model predicts the user's score for an item based on the known relationship between the user and the user's score for the item, and generates recommendations based on such scores.
The social recommendation model may be expressed in the form of:
social recommendation model ═ recommendation model + social model formula (1)
Referring to FIG. 2, a schematic diagram of a single-view social recommendation model is shown. As will be described more fully below.
The recommendation model and the social model are described separately below.
Recommendation model
A recommendation model described herein (e.g., recommendation model 202 shown in fig. 2) is a machine learning model that recommends items to a user. The item recommended to the user may be a commodity or other items such as content.
Various recommendation models may be employed. One common recommendation model is the latent semantic model. Matrix decomposition models are the most widespread implementation of latent semantic models. The matrix decomposition model will be described below as an example.
The core assumption of the matrix factorization based recommendation algorithm is to represent users and items with potential vectors. These potential vectors represent features that are common to a portion of the user and the item. For example, a potential vector may appear to a user as a user preference feature; for an item, a potential vector may be characterized as an item attribute. In many cases, a potential vector may not have a practical meaning nor must it have very good interpretability, nor have each dimension have a definite label name, and is therefore referred to as a "potential vector". Through matrix decomposition, two small matrixes containing potential vectors are obtained, one is used for representing implicit characteristics of users, the other is used for representing implicit characteristics of articles, and element values of the matrixes represent the conformity degree of corresponding users or articles to various implicit factors, and the matrixes have positive and negative effects.
Assume that there is a set of users u and a set of items v. There is also a scoring matrix R of the items by the user, where each element R in the matrix R
ijRepresenting user u
iFor article v
jThe score of (1). The scoring matrix may be decomposed into a user-latent factor matrix and an item-latent factor matrix using a matrix decomposition algorithm. Suppose using U
iAnd V
jRespectively represent users u
iAnd an article v
jThe simplified recommendation model's objective function can be represented as follows:
the objective function described above is intended to minimize the error of the true score from the predicted score.
It can be appreciated that the objective function is a simple example, and other objective functions may be adopted, for example, adding a regular term, which is not described herein again.
Equation 2 can be solved using methods such as gradient descent. For example, U and V can be obtained by iteration using a gradient descent method. Subsequently, for a particular user, user information for that particular user may be obtained, such as known rating information for various items by the user, and social information of the user with the user. The candidate items may then be scored using the found U and V based on user information. Based on the scores for each candidate item, a recommendation for the user may be generated.
Other recommendation models may also be employed. For example, examples of recommendation models that may be employed include, but are not limited to: domain-based collaborative filtering models, other latent semantic models such as latent semantic analysis, topic models, and the like. Preferably, the recommendation model employed in the present specification is a user-item relationship based recommendation model or a user-based recommendation model.
Single-view social model
Referring to FIG. 1A, a prior art social graph with a single view is shown. As shown in FIG. 1A, the social graph has only one view, representing only a single type of social relationship (e.g., friendship). For example, in FIG. 1A, user 102 is friend with users 106 and 110, and user 104 is friend with users 106 and 108. It is noted that although shown in FIG. 1A as a two-way relationship, one skilled in the art will appreciate that the social relationship may have a one-way relationship (e.g., one-way attention, one-way money transfers, etc.).
In some examples, the social relationship may have a strength. For example, where the relationship between the user 102 and the user 106 is a friendship, the strength of the social relationship may be based on the degree of affinity of the user 102 and the user 106. For example, the affinity may be determined based on the similarity of the users 102 and 106, the frequency of interaction, greetings used, and the like.
For another example, where the relationship between the user 102 and the user 106 is an affinity relationship, the strength of the social relationship may be based on the affinity relationship between the user 102 and the user 106. For example, a parent-child relationship may be considered to be more robust than a tertiary-nephew relationship.
As another example, where the relationship between the user 102 and the user 106 is a transfer relationship, the strength of the social relationship may be based on the frequency of transfers between the user 102 and the user 106, the amount of transfers, and the like.
It can be understood that the stronger the relationship between two users, the better the recommendation to one of the users based on the related information of the other user (e.g. the preference of the user for the item, etc.).
In the above example, the strength of the relationship between users is explained as an example. One skilled in the art will appreciate that social relationships may also have other dimensions, such as a degree of similarity between two users. Two similar users may have similar preferences for items so that social network-based recommendations may be made using a social model that reflects the similarity between users.
To describe such a single-view social graph, there are many social models (e.g., the single-view social model 204 shown in FIG. 2 or the single-view social models 312, 314 … … 316 shown in FIG. 3). For example, suppose a user-user social relationship matrix is represented by S, each element S of the matrix S
ifRepresenting user u
iWith user u
fThe strength and/or similarity of social relationships, the objective function of one type of social model may be represented as follows:
social models that employ other constraints also exist. For example, in another social model, user u may be presented with
iWith user u
fThe average relationship between them is constrained and the objective function of such a model can be expressed as follows:
these single-view social models are well known to those skilled in the art and will not be described in detail herein.
Taking the above example as an example, based on known strength of relationship and/or similarity of users in the single-view social model, the unknown strength of relationship and/or similarity of two users may be predicted.
Single-view social recommendation model
Referring to FIG. 2, a schematic diagram of a single-view social recommendation model 206 is shown. As shown in FIG. 2, a single-view social recommendation model 206 may be implemented by combining a recommendation model (based on user-item relationships) 202 and a single-view social model (based on user-user relationships) 204.
Through the single-view social recommendation model, the relation between the user and the item is considered when the recommendation is generated, and the relation between the user and the user in a single view is also considered, so that compared with a simple recommendation model, the single-view social recommendation model provides a new dimension, and better recommendation is realized.
Continuing with the example above, based on equations 1,2, and 3 above, an objective function that may determine a single view social recommendation model may be represented as follows:
the single-view social recommendation model can also use a gradient descent method to solve the objective function through repeated iteration and continuous updating so as to train the single-view social recommendation model. The trained single-view social recommendation model may be used to provide recommendations for new users.
Multi-view social model
Referring to FIG. 1B, a social graph having multiple views is shown. Unlike the social graph of FIG. 1A, which has a single view, the social graph shown in FIG. 1B has multiple views, e.g., each social graph may correspond to a social relationship type. Preferably, the different views correspond to different social relationship types. For example, it is possible that one view corresponds to a friendship in a certain social network, another view corresponds to a transfer relationship, yet another view corresponds to a communication relationship, and so on. Preferably, the multiple views are from more than one social networking service. In some examples, one or more of the multiple views are from the same social networking service. For example, a pay may provide a view associated with a friendship and a view associated with a transfer relationship at the same time. In other examples, each view of the plurality of views is from a different social networking service. For example, the communication relationship may be obtained from a phone book, the payment relationship may be obtained from a payment instrument, and so on. Preferably, the social graph 1B constitutes a heterogeneous network.
For example, as shown in FIG. 1B, user 102 has not only a friendship, but also a transferring relationship with users 106 and 110, but only a transferring relationship with user 102 and user 108. As another example, as shown in FIG. 1B, users 104 and 106 have only a friendship, users 104 and 108 have only a communication, and single users 104 and 106 have both a friendship and a communication. Likewise, although both shown in FIG. 1B as bi-directional relationships, one skilled in the art will appreciate that the social relationships may have uni-directional relationships (e.g., uni-directional attention, uni-directional transfers, etc.).
Similar to FIG. 1A, the various types of social relationships shown in FIG. 1B may all have strength. Examples of the strength of social relationships have been described above with reference to FIG. 1A and are not described in detail herein.
It will be appreciated that the multi-view social graph of FIG. 1B contains more information than the single-view social graph of FIG. 1A. For example, in the example of FIG. 1B, assuming that the strength of the communication relationship between users 104 and 108 is the same as the strength of the communication relationship between users 104 and 110, the additional friendship between users 104 and 108 may mean that users 104 are more intimate with users 108 than with users 110. Of course, this is only a simple example, and one skilled in the art can mine more complex information based on multiple views of the social graph.
Likewise, in addition to the strength of the relationship between users, the relationship between users may have other dimensions, such as similarity between users, and so forth.
To leverage this richer information, the present application proposes a multi-view social model to describe a multi-view social graph. In embodiments of the present invention, a multi-view social model (e.g., multi-view social model 304 of FIG. 3) is constructed by combining multiple single-view social models (e.g., single-view social model 1312, single-view social model 2314, single-view social model … …, single-view social model M316 of FIG. 3) using a combining function. Preferably, each single-view social model may correspond to a social relationship type. Preferably, the different single-view social models may correspond to the same or different types of social relationship, but the multiple single-view social models correspond to at least two types of social relationship. The plurality of social relationship types may include, for example, two or more of a friendship, an affinity, a correspondence, and a transfer relationship. Preferably, the plurality of single-view social models correspond to a plurality of social networking services. Preferably, each of the plurality of single-view social models is based on a different social networking service. Alternatively, two or more of the multiple single-view social models are based on the same social networking service (e.g., a pay for your social networking service may provide both transfer and friend types of social relationship, and thus may be used to generate two single-view social models).
For example, assuming that M single-view social models, each representing a type of social relationship, are represented by M e {1,2, …, M }, the objective function of the multi-view social model may be represented as follows:
wherein
Representing the strength of the social relationship represented by the single-view social model m between user i and user f and/or the similarity of the two users.
Representing a combined function
To combine multiple single-view social models m e {1,2, … }. Said combination function
There may be various functions, i.e., the multiple single-view social models may be combined in a variety of different ways.
In one embodiment of the invention, the functions are combined
May be a summation function
It will be appreciated that while the above combining function employs a general summing function, a weighted summing function may also be employed to assign different weight values to different single-view social models. For example, certain social network views may be given a higher weight value in cases where they are more valuable to recommend than others.
In another embodiment of the invention, the functions are combined
May be an averaging function
In yet another embodiment of the present invention, the functions are combined
May be a pooling function
The pooling function is a common function in convolutional networks and its meaning and implementation is understood by those skilled in the art and will not be described in detail here.
It will be appreciated by those skilled in the art that other combining functions may also be employed to combine the multiple views.
Which type of combinatorial function is specifically used can be chosen by the developer as desired.
It is noted that although in the above example, each view is represented by the same single-view social model (e.g., a matrix decomposition model), it can be appreciated that different social models can be employed for different views. In this case, multiple single-view social models may still be combined in the manner as exemplified above. For example, a plurality of different single-view social models may be summed, weighted summed, averaged, pooled, and the like.
Multi-view social recommendation model
In order to generate recommendations by fully utilizing richer information provided by the multi-view social graph provided by the invention, so that the accuracy of the recommendations is improved, the invention provides a multi-view social recommendation model.
Referring to FIG. 3, a schematic diagram of a multi-view social recommendation model is shown, according to an embodiment of the present description. As shown in fig. 3, by combining the recommendation model (based on user-item relationships) and the multi-view social model (based on multiple types of user-user relationships), various different user relationships can be fully utilized, barriers between different types of social relationships are broken, the number of dimensions utilized is further increased, and thus a better recommendation effect is provided compared with the single-view social recommendation model.
In particular, the multi-view social recommendation model 306 combines the recommendation model 302 with the multi-view social model 304 to generate recommendations for the user.
Continuing with the above example, replacing the social model in equation (1) with a multi-view social model, a formula for a multi-view social recommendation model can be obtained:
multi-view social recommendation model ═ recommendation model + multi-view social model formula (10)
Taking the example objective function above as an example, based on equation 10, equation 2, and equation 6 above, the objective function that may determine the multi-view social recommendation model may be represented as follows:
the above combination function
May be the averaging function in equation 7 above
Summation function
Pooling function
And so on.
As already explained above, the specific choice of which combination function is to be selected can be chosen by the developer as desired.
Generating recommendations based on a multi-view social recommendation model
After the multi-view social recommendation model is obtained, the multi-view social recommendation model can be trained by using training data, and the multi-view social recommendation model is updated iteratively. After training the multi-view social recommendation model, recommendations may be generated for the user with the trained social recommendation model.
Taking the above example objective function as an example, for the objective function of the multi-view social recommendation model introduced above, a gradient descent method may be used for solving. For example, for the objective function shown in equation 11, a derivative may be obtained, and by updating iteration, U and V may be obtained, i.e. parameters of the multi-view social recommendation model may be obtained. Using the parameter, based on the user potential vector for the particular user and the item potential vector for the particular item, U is calculated according to the formula
TV may predict the rating of the particular item by the particular user. Recommendations may then be made based on the rating of the item in the candidate set of items by the particular user. For example, the highest scoring item may be recommended to the user. Alternatively, the user may be presented withRecommending a plurality of items scored in the top. For example, the top three ranked items may be recommended to the user.
It should be noted that although the article (commodity) is exemplified in the above example, it should be understood that what is recommended may be a commodity (e.g., a physical commodity or a virtual commodity) or content (e.g., audio, video, microblog posts, etc.).
Multi-view social recommendation method
Referring to FIG. 4, a flow diagram of a method 400 for generating recommendations for a user is shown, in accordance with an embodiment of the present invention.
The method 400 may include: at step 402, a recommendation model is generated for recommending items for a user. As already described in detail above, the recommendation model may employ any suitable known recommendation model, such as a matrix decomposition model.
The method 400 may also include: at step 404, a multi-view social model is generated, the multi-view social model corresponding to a plurality of social relationship types. Specifically, generating the multi-view social model may include: a plurality of single-view social models are generated, where each single-view social model may correspond to a social relationship type. Preferably, the different single-view social models may correspond to the same or different types of social relationship, but the multiple single-view social models correspond to at least two types of social relationship. As described above, the plurality of single-view social models may correspond to a plurality of social networking services. Preferably, each of the plurality of single-view social models corresponds to a different social networking service. Alternatively, two or more of the multiple single-view social models may be based on the same social networking service. Preferably, the plurality of social relationship types includes two or more of a friendship, an affinity, a correspondence, and a transfer relationship.
Generating the multi-view social model may include: the method can also comprise the following steps: combining the plurality of single-view social models to arrive at a multi-view social model.
The single-view social models may be combined according to a combining function. For example, combining the plurality of single-view social models may include summing the objective functions of the plurality of single-view social models (e.g., equation 7 above), summing by weights, averaging (e.g., equation 8 above), performing a pooling operation (e.g., equation 9 above). Combining the plurality of single-view social models may include determining a combining function for combining the plurality of single-view social models. Preferably, a selection of a combining function for combining the single-view social models may be received from the developer.
The method 400 may also include: at step 406, the recommendation model is combined with the multi-view social model to arrive at a multi-view social recommendation model. The manner in which the recommendation model and the multi-view social model are combined may be the same as, for example, the manner in which the recommendation model and the single-view social model are combined.
Optionally, the method 400 may further include: at step 408, the multi-view social recommendation model is trained. Training the multi-view social recommendation model may include, for example: iteratively solving an objective function of the multi-view social recommendation model using a gradient descent method to determine parameters of the multi-view social recommendation model. This parameter may later be used to determine a score for the candidate item to be recommended.
The method 400 may also include: at step 410, recommendations are generated for the user using the trained multi-view social recommendation model.
This step may include: the method comprises the steps of obtaining user information of a user, wherein the user information comprises known scoring information of the user on an article and social information of a plurality of social relation types between the user and the user. The user information may be represented, for example, as a vector representation of the user, such as a one-hot vector, or the like.
This step may further include: determining a score for each of a plurality of candidate items using a trained multi-view social recommendation model based on the user information of the user. For example, based on the parameters of the multi-view social recommendation model, the user's score for the item may be solved.
This step may further include: recommending items to the user based on the scores of the plurality of candidate items. For example, the highest scoring item may be recommended to the user. Alternatively, a plurality of items scored top may be recommended to the user. For example, the top three ranked items may be recommended to the user. The recommendations may include, but are not limited to, recommendations for merchandise or recommendations for content, etc.
Generating recommendations using a trained recommendation model may be performed in any manner known to those skilled in the art, and the process is not described in detail herein.
Multi-view social recommendation system
Referring to FIG. 5, a schematic diagram of a system 500 for generating recommendations for a user is shown, in accordance with an embodiment of the present invention.
As shown in fig. 5, system 500 may include a user information acquisition module 502. The user information obtaining module 502 may obtain user information including known rating information of the user for the item and social information of a plurality of social relationship types of the user and the user. The user information may be, for example, a vector representation of the user obtained based on the user ID.
The system 500 may also include a multi-view social recommendation module 506. The multi-view social recommendation module 506 may include a recommendation module 508 and a multi-view social module 510. The recommendation module 508 may implement the recommendation model 302 as described above. The multi-view social recommendation module 510 may include a plurality of single-view social modules 512 and a combination module 514. Each single-view social module 512 may implement a single-view social model. The combining module 514 is used to combine the single-view social modules 512 into the multi-view social module 510, for example, combining may be performed using any of a variety of combining functions as described above. The multi-view social recommendation module 506 may generate recommendations for the user based on the user information from the user information acquisition module 502.
The present application also discloses a computer-readable storage medium comprising computer-executable instructions stored thereon that, when executed by a processor, may cause the processor to perform the methods of the embodiments described herein.
The present application also discloses a system that may include means for performing the methods of the various embodiments described herein.
It is to be understood that methods according to embodiments of the present invention may be implemented in software, firmware, or a combination thereof.
It is to be understood that the specific order or hierarchy of steps in the methods disclosed is an illustration of exemplary processes. Based upon design preferences, it is understood that the specific order or hierarchy of steps in the methods may be rearranged. The accompanying method claims present elements of the various steps in a sample order, and are not meant to be limited to the specific order or hierarchy presented unless specifically recited herein.
It should be understood that an element described herein in the singular or shown in the figures only represents that the element is limited in number to one. Furthermore, modules or elements described or illustrated herein as separate may be combined into a single module or element, and modules or elements described or illustrated herein as single may be split into multiple modules or elements.
It is also to be understood that the phraseology and terminology employed herein are for the purpose of description and that the invention is not to be regarded as limited to such terminology and terminology. The use of such terms and expressions is not intended to exclude any equivalents of the features shown and described (or portions thereof), and it is recognized that various modifications may be made within the scope of the claims. Other modifications, variations, and alternatives are also possible. Accordingly, the claims should be looked to in order to cover all such equivalents.
Also, it should be noted that although the present invention has been described with reference to the current specific embodiments, it should be understood by those skilled in the art that the above embodiments are merely illustrative of the present invention, and various equivalent changes or substitutions may be made without departing from the spirit of the present invention, and therefore, it is intended that all changes and modifications to the above embodiments be included within the scope of the claims of the present application.
Claims (15)
1. A method for generating recommendations for a user, the method comprising:
generating a recommendation model, wherein the recommendation model is used for recommending articles for a user;
generating a multi-view social model, the multi-view social model corresponding to a plurality of social relationship types;
combining the recommendation model with the multi-view social model to arrive at a multi-view social recommendation model;
training the multi-view social recommendation model; and
generating recommendations for the user using a trained multi-view social recommendation model.
2. The method of claim 1, wherein generating the multi-view social model comprises:
generating a plurality of single-view social models, each single-view social model corresponding to a social relationship type;
combining the plurality of single-view social models to arrive at a multi-view social model.
3. The method of claim 2, wherein the plurality of single-view social models correspond to a plurality of social networking services.
4. The method of claim 2, wherein each of the plurality of single-view social models corresponds to a different social networking service.
5. The method of claim 1, wherein the plurality of social relationship types comprise two or more of a friendship, an affinity, a correspondence, and a transfer relationship.
6. The method of claim 2, wherein combining the plurality of single-view social models comprises summing, weighted summing, averaging, or performing a pooling operation on objective functions of the plurality of single-view social models.
7. The method of claim 1, wherein combining the plurality of single-view social models comprises determining a combining function for combining the plurality of single-view social models.
8. The method of claim 1, wherein generating a recommendation for the user using a trained multi-view social recommendation model comprises:
acquiring user information of the user, wherein the user information comprises known scoring information of the user on an article and social information of a plurality of social relation types between the user and the user;
determining a score for a plurality of candidate items using a trained multi-view social recommendation model based on user information of the user; and
recommending items to the user based on the scores of the plurality of candidate items.
9. The method of claim 1, wherein the recommendation model is a matrix decomposition model.
10. The method of claim 1, wherein training the multi-view social recommendation model comprises:
iteratively solving an objective function of the multi-view social recommendation model using a gradient descent method to determine parameters of the multi-view social recommendation model.
11. The method of claim 10, wherein generating a recommendation for the user using a trained multi-view social recommendation model comprises:
a score for the candidate item is determined using the determined parameters.
12. The method of claim 1, wherein generating recommendations for the user comprises generating recommended goods or recommended content for the user.
13. A system for generating recommendations for a user, the system comprising:
the system comprises a user information acquisition module, a user information acquisition module and a user information processing module, wherein the user information acquisition module is used for acquiring user information of a user, and the user information comprises known scoring information of the user on an article and social information of a plurality of social relationship types between the user and the user;
a multi-view social recommendation module to determine scores for a plurality of candidate items based on user information of the user, the multi-view social recommendation module comprising a recommendation module and a multi-view social module, wherein:
the recommendation module implements a recommendation model for recommending items for a user,
the multi-view social recommendation module includes a plurality of single-view social modules that implement a plurality of single-view social models corresponding to a plurality of social relationship types, and a combining module to combine the plurality of single-view social modules into the multi-view social module.
14. A computer-readable storage medium storing instructions that, when executed by a computer, cause the computer to perform the method of any of claims 1-12.
15. A system comprising means for performing the method of any one of claims 1-12.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910994809.3A CN110781389A (en) | 2019-10-18 | 2019-10-18 | Method and system for generating recommendations for a user |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910994809.3A CN110781389A (en) | 2019-10-18 | 2019-10-18 | Method and system for generating recommendations for a user |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110781389A true CN110781389A (en) | 2020-02-11 |
Family
ID=69385999
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910994809.3A Pending CN110781389A (en) | 2019-10-18 | 2019-10-18 | Method and system for generating recommendations for a user |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110781389A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022262561A1 (en) * | 2021-06-17 | 2022-12-22 | 腾讯科技（深圳）有限公司 | Multimedia resource processing method and apparatus, and device and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105653535A (en) * | 2014-11-13 | 2016-06-08 | 中国科学院沈阳计算技术研究所有限公司 | Media resource recommendation method |
| CN105956093A (en) * | 2016-04-29 | 2016-09-21 | 浙江大学 | Individual recommending method based on multi-view anchor graph Hash technology |
| CN108346067A (en) * | 2018-01-23 | 2018-07-31 | 中国科学技术大学苏州研究院 | Social networks advertisement sending method based on natural language processing |
| CN110245301A (en) * | 2018-11-29 | 2019-09-17 | 腾讯科技（深圳）有限公司 | A kind of recommended method, device and storage medium |
-
2019
- 2019-10-18 CN CN201910994809.3A patent/CN110781389A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105653535A (en) * | 2014-11-13 | 2016-06-08 | 中国科学院沈阳计算技术研究所有限公司 | Media resource recommendation method |
| CN105956093A (en) * | 2016-04-29 | 2016-09-21 | 浙江大学 | Individual recommending method based on multi-view anchor graph Hash technology |
| CN108346067A (en) * | 2018-01-23 | 2018-07-31 | 中国科学技术大学苏州研究院 | Social networks advertisement sending method based on natural language processing |
| CN110245301A (en) * | 2018-11-29 | 2019-09-17 | 腾讯科技（深圳）有限公司 | A kind of recommended method, device and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 胡吉明: "《社会网络环境下基于用户关系的信息推荐服务研究》", 31 March 2015 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022262561A1 (en) * | 2021-06-17 | 2022-12-22 | 腾讯科技（深圳）有限公司 | Multimedia resource processing method and apparatus, and device and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111061946B (en) | Method, device, electronic equipment and storage medium for recommending scenerized content | |
| US11734717B2 (en) | Dynamic predictive similarity grouping based on vectorization of merchant data | |
| CN110162693A (en) | A kind of method and server of information recommendation | |
| CN111259263B (en) | Article recommendation method and device, computer equipment and storage medium | |
| CN107688605B (en) | Cross-platform data matching process, device, computer equipment and storage medium | |
| CN109087178A (en) | Method of Commodity Recommendation and device | |
| CN110209922A (en) | Object recommendation method, apparatus, storage medium and computer equipment | |
| CN102576438A (en) | Method and apparatus for executing a recommendation | |
| US9489638B2 (en) | Method and apparatus for propagating user preference information in a communications network | |
| CN110413888B (en) | Book recommendation method and device | |
| Gao et al. | Mining human mobility in location-based social networks | |
| Díez et al. | Towards explainable personalized recommendations by learning from users’ photos | |
| CN110008397A (en) | A kind of recommended models training method and device | |
| CN110019650A (en) | Method, apparatus, storage medium and the electronic equipment of search associational word are provided | |
| US11783436B2 (en) | Magellan: a context-aware itinerary recommendation system built only using card-transaction data | |
| CN111611499A (en) | Collaborative filtering method, collaborative filtering device and collaborative filtering system | |
| Kim et al. | Improving a recommender system by collective matrix factorization with tag information | |
| CN113656699B (en) | User feature vector determining method, related equipment and medium | |
| CN110929172A (en) | Information selection method and device, electronic equipment and readable storage medium | |
| CN110781389A (en) | Method and system for generating recommendations for a user | |
| CN111353001B (en) | Method and device for classifying users | |
| CN112465593A (en) | Method for realizing fashion suit recommendation through graph neural network | |
| CN116484092A (en) | Hierarchical attention network sequence recommendation method based on long-short-term preference of user | |
| CN113327154B (en) | E-commerce user message pushing method and system based on big data | |
| CN110928920B (en) | Knowledge recommendation method, system and storage medium based on improved position social contact |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200211 |
|
| RJ01 | Rejection of invention patent application after publication |