CN111026973B - Commodity interest degree prediction method and device and electronic equipment - Google Patents
Commodity interest degree prediction method and device and electronic equipment Download PDFInfo
- Publication number
- CN111026973B CN111026973B CN201911375799.1A CN201911375799A CN111026973B CN 111026973 B CN111026973 B CN 111026973B CN 201911375799 A CN201911375799 A CN 201911375799A CN 111026973 B CN111026973 B CN 111026973B
- Authority
- CN
- China
- Prior art keywords
- user
- commodity
- users
- relationship network
- interestingness
- 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.)
- Active
Links
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
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- 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
- 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
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
Abstract
The invention provides a method and a device for predicting commodity interestingness and electronic equipment; the method comprises the following steps: acquiring a user relationship network diagram; the user relationship network graph comprises: the system comprises user nodes representing users and edges connecting the user nodes to represent that friend relationships exist among the users; wherein, part of the user nodes are associated with labels for expressing commodity interestingness; inputting the user relationship network graph into a graph attention network model to obtain the influence degree between the two user nodes connected with the edge; generating a weighted user relation network graph according to the influence degree and the user relation network graph; and obtaining a commodity interest degree prediction result by using a preset prediction algorithm according to the weighted user relationship network graph.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a commodity interestingness prediction method and device and electronic equipment.
Background
With the advent of the network era, electronic commerce has been rapidly developed, and the number of goods provided by electronic commerce operators has rapidly increased. How to efficiently provide services for the full-fledged Linglan commodities based on the interest of users in the commodities is the key for electronic commerce operators to realize accurate personalized services and improve the service quality. Specifically, the commodity interestingness prediction of the user can be widely applied to searching, recommending and advertising, and more accurate service can be provided for the user by utilizing the commodity interestingness of the user. Currently, there is a need to provide a scheme with higher prediction accuracy.
Disclosure of Invention
In view of the above, the present invention provides a method and an apparatus for predicting interest level of a product, and an electronic device.
Based on the above purpose, the present invention provides a method for predicting interest level of a commodity, comprising:
acquiring a user relationship network diagram; the user relationship network graph comprises: the system comprises user nodes representing users and edges connecting the user nodes to represent that friend relationships exist among the users; wherein, part of the user nodes are associated with labels for expressing commodity interestingness;
inputting the user relationship network graph into a graph attention network model to obtain the influence degree between two user nodes connected with the edges;
according to the influence degree, assigning a weight to the edge in the user relationship network graph to obtain a weighted user relationship network graph;
and predicting to obtain a commodity interest degree prediction result according to the weighted user relationship network graph.
In another aspect, the present invention further provides a device for predicting interest level of a commodity, including:
the acquisition module is configured to acquire a user relationship network diagram; the user relationship network graph comprises: the system comprises user nodes representing users and edges connecting the user nodes to represent that friend relationships exist among the users; wherein, part of the user nodes are associated with labels for expressing commodity interestingness;
the influence degree determining module is configured to input the user relationship network graph into a graph attention network model to obtain the influence degree between the two user nodes connected with the edges;
the weighting module is configured to endow the edges in the user relationship network graph with weights according to the influence degree to obtain a weighted user relationship network graph;
and the prediction module is configured to predict a commodity interest degree prediction result according to the weighted user relationship network diagram.
In yet another aspect, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor implements the method according to any one of the above aspects when executing the program.
From the above, according to the commodity interest degree prediction method, the commodity interest degree prediction device and the electronic equipment, the users with friend relations are represented through the user relation network graph, the influence degrees among the user nodes are obtained through the attention network model, the user relation network graph is weighted according to the influence degrees, the influence degrees among the users are reflected in the user relation network graph, finally, the commodity interest degree prediction result is obtained through the prediction algorithm and the weighted user relation network graph, and the commodity interest degree prediction accuracy is effectively improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
FIG. 1 is a flowchart of a method for predicting interest level of a commodity according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of a user relationship network in an embodiment of the invention;
FIG. 3 is a flowchart illustrating steps for generating a user relationship network diagram according to an embodiment of the present invention;
FIG. 4 is a flowchart illustrating steps for recommending information of a product according to a prediction result of interest degree of the product according to an embodiment of the present invention;
FIG. 5 is a schematic structural diagram of a product interest prediction apparatus according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to specific embodiments and the accompanying drawings.
As described in the background section, when a user purchases a commodity (which may be a physical commodity or a virtual service) through an e-commerce platform on the internet, the commodity interest level of the user in the commodity can be reflected. In the related art, if a certain user purchases a certain product, the product interest of the user in the product (or the like) may be defined as 100%.
In addition, in various service platforms providing different service services, there is a friend relationship between users, such as a pay friend and a nail friend. For a user with a friend relationship, if many friends of a user are interested in a certain commodity, the user can be considered to be interested in the certain commodity to some extent.
Based on the above, in the related art, there is a technology for predicting the interest level of a product by using the interest level of the product known by some users and other users having a friend relationship with the users. However, the above related technologies only consider the friend relationship among users simply, that is, consider it as a simple corresponding relationship; however, in practical application, the expression of the friend relationship between users is more complicated, and the related technology cannot be consistent with the practical application situation, so that the problem of insufficient accuracy of commodity interestingness prediction is caused.
In view of the above problems, one or more embodiments of the present specification provide a method, an apparatus, and an electronic device for predicting interest level of a commodity. On the basis of considering the friend relationship among users, the influence degree among the users with the friend relationship is further considered. That is, the relationship is also friend relationship, but the degrees of friend relationship are different, some are closer and some are farther; correspondingly, the friend relationship with closer relationship has larger influence on the user, and the friend relationship with farther relationship has smaller influence on the user. By considering the influence degree among users, the scheme of one or more embodiments of the specification is more suitable for the actual application situation, and the accuracy of commodity interest degree prediction is effectively improved.
The technical solutions provided by the embodiments of the present disclosure are described in detail below with reference to the accompanying drawings.
Referring to fig. 2, the method for predicting interest level of a commodity according to the present embodiment includes the following steps:
In this embodiment, a network graph is applied, where the network graph is a data structure, and is composed of nodes and undirected edges connecting the nodes, and is used to represent a plurality of objects having a certain relationship in an actual application scene; the nodes are expressed as vectors, and data used in the process of establishing the network graph are included in the nodes in a form of corresponding vector features.
In this embodiment, the user relationship network diagram is used to represent a plurality of users having a friend relationship. Specifically, the user relationship network graph comprises a plurality of user nodes representing users. For some user nodes, an edge is connected between the user nodes, and the edge indicates that a friend relationship exists between the two user nodes connected with the edge. In this embodiment, the friend relationship may be a friend relationship of a friend system based on some service platforms, such as a pay friend, a nail friend, and a wechat friend; the friend relationship may also be obtained through analysis or prediction using other related techniques, such as prediction through a machine learning model from interaction data between users.
Referring to fig. 2, an example of the user relationship network diagram of the present embodiment includes: five USER nodes of USER _ A, USER _ B, USER _ C, USER _ D and USER _ E. The USER corresponding to the USER node USER _ A has friend relationship with the USERs corresponding to the USER nodes USER _ B, USER _ C and USER _ D; the USER corresponding to the USER node USER _ B has a friend relationship with the USERs corresponding to the USER node USER _ a and the USER node USER _ E. Those skilled in the art will appreciate that the user relationship network diagram shown in fig. 2 is merely an example for convenience of explanation, and in an actual implementation, the user relationship network diagram may include more nodes and edges.
In addition, some user nodes in the user relationship network graph are associated with tags for indicating commodity interestingness, and the tags are generally associated based on existing data. And the transaction record data of the E-commerce platform indicates that a certain user purchases a certain commodity through the transaction record data, the user can be determined to have interest in the commodity, and the user node representing the user is the corresponding label which is associated to represent the interest degree of the commodity. In addition, the user nodes may also be associated with tags based on the results of other related commodity interestingness determination techniques.
Specifically, in the data structure, a label associated with a user node constitutes a so-called example in machine learning. The user nodes are represented by vectors and the labels are represented as a scalar value. In this embodiment, the associated label is assigned as 1 corresponding to the user node known to have commodity interest in the commodity; the other user nodes have an associated label value of 0. In the training process, vectors of all user nodes included in the network graph are used as a sample set, and a set of all labels is input into the graph attention network model together for training, the trained graph attention network model can establish a mapping relation between the user nodes and the labels, namely, for any user node, the corresponding label can be obtained, and the obtained label takes the value from the interval of [0-1 ]. It can be seen that, since some user nodes in the user relationship network graph are associated with tags, the user relationship network graph of the embodiment forms a supervised network graph data as a whole. In this embodiment, the user relationship network graph may be generated by other external programs or devices, and when the method of this embodiment is implemented, the user relationship network graph is directly obtained; or, when the method of this embodiment is implemented, the transaction data including the friend relationship between users and the user is obtained, and then the data is generated.
And 102, inputting the user relationship network graph into a graph attention network model to obtain the influence degree between the two user nodes connected with the edges.
In this embodiment, a Graph Attention network model (GAN) is used. The graph attention network model is a classical graph neural network structure, the input of the graph attention network model is a network graph, the output of the graph attention network model is vector expression of each node in the network graph, and the difference of the graph attention network model from the traditional graph neural network structure is that the graph attention network model can calculate influence among the nodes through an attention mechanism so as to reflect different degrees of mutual influence among different nodes. The specific principle and operation of the graph attention network model are the prior art, and the detailed description is omitted in the embodiment.
In this embodiment, the user relationship network diagram obtained in the foregoing step is input into the graph attention network model. And outputting the influence degree between two user nodes connected with edges in each group in the user relationship network graph through the calculation processing of the graph attention network model. Besides, the graph attention network model outputs the influence degree, and also outputs the representation vectors of each node in the user medium network and relevant model parameters of other graph attention network models.
Referring to fig. 2, the numbers marked on the edges of the graph of the user relationship network are to note the influence of the output of the force network model. For example, the degree of influence between the USER node USER _ a and the USER node USER _ B is 0.61, the degree of influence between the USER node USER _ C is 0.8, and the degree of influence between the USER node USER _ D is 0.77.
And 103, giving a weight to the edge in the user relationship network graph according to the influence degree to obtain a weighted user relationship network graph.
In this embodiment, the user relationship network graph is weighted according to the influence between the user nodes obtained in the foregoing steps. The specific weighting process is to use the influence between two user nodes as the weight of the edge connecting the two user nodes. And after weighting processing, obtaining a weighted user relationship network graph.
For a general user relationship network graph, the weight of each edge is not specially set, and for a user node, the weight of each edge connected with the user node is evenly distributed. In this embodiment, the weight of each edge in the user relationship network graph is set by noting the influence degree calculated by the force network model, so as to reflect the mutual influence between user nodes, and specifically, the influence of different friends on a user in different degrees can be reflected, so that the weighted user relationship network graph of this embodiment can better conform to the actual friend relationship between users, and accordingly, the accuracy of subsequent prediction can be improved.
And step 104, predicting to obtain a commodity interest degree prediction result according to the weighted user relation network graph.
In this embodiment, prediction is performed using a prediction algorithm based on the obtained weighted user relationship network graph. In this embodiment, a Label Propagation Algorithm (LPA) is taken as an example to perform prediction. The label propagation algorithm is a graph-based semi-supervised learning method, and the basic idea is to use label information of labeled nodes to predict label information of unlabeled nodes. In the embodiment, some user nodes in the weighted user relationship network graph are associated with tags to form a supervised network graph. And predicting through a label propagation algorithm to obtain a commodity interest degree prediction result. In this embodiment, the product interest prediction result includes the product interest of the user node not associated with the label in the weighted user relationship network diagram. In this embodiment, if the label assignment of the user node associated with the label is 1, the commodity interest level of the output user node is a numerical value between 0 and 1. According to the commodity interest degree prediction result, the interest degree of the user for purchasing commodities, which is represented by the user node, can be judged.
Further, the result of predicting the interest degree of the commodity may be used as input data. Inputting the data into each functional system of the online service platform to realize richer functions. For example, inputting the commodity interest degree prediction result into a commodity recommendation system, and recommending corresponding commodity information to the user; or inputting the commodity interest degree prediction result into a software page system, and realizing personalized page setting for the user according to the commodity interest degree.
In addition, according to different implementation requirements, other prediction algorithms such as the global connectivity search Algorithm (CPM), the overlapping Community discovery Algorithm based on tag Propagation (COPRA), the Balanced Multi-tag Propagation Algorithm (BMLPA), and the Community structure discovery Algorithm (Girvan-Newman Algorithm, GN) may be used for prediction.
It can be seen that, in the commodity interest degree prediction method of this embodiment, users with friend relationships are represented through the user relationship network graph, the influence degree between user nodes is obtained through the graph attention network model, the user relationship network graph is weighted according to the influence degree, so that the influence degree between users is reflected in the user relationship network graph, and finally, a commodity interest degree prediction result is obtained through a prediction algorithm in combination with the weighted user relationship network graph, thereby effectively improving the accuracy of commodity interest degree prediction.
As an alternative embodiment, referring to fig. 3, the method for predicting commodity interestingness further includes a step of generating a user relationship network diagram, specifically:
In this embodiment, the user relationship network diagram is generated when the method of this embodiment is executed. Specifically, user friend relationship data used for generating a user relationship network diagram is obtained. The user friend relation data can be obtained from a server system of a service platform, software and the like which provide friend functions, for example, the user friend relation data recorded with friend relations among users of the payment treasures is obtained from the payment treasures. According to the friend relation data of the users, a plurality of users with friend relations can be determined.
In this embodiment, historical transaction data is obtained from the e-commerce platform. And determining the purchase records of the users determined according to the friend relationship data of the users for a certain commodity from the historical transaction data, thereby determining that the users having the purchase records of the commodity have commodity interest in the commodity.
In this embodiment, according to a plurality of users having a friend relationship determined by the friend relationship data of the users, the nodes represent the users, and the edges represent the friend relationship between the users, so as to generate a user relationship network graph including the user nodes and the edges.
In this embodiment, according to the commodity interestingness of some users for the commodity determined in the foregoing steps, the users in the user relationship network graph are associated with the tags used for representing the commodity interestingness.
For specific contents of the user relationship network graph and the label generated in this embodiment, reference may be made to the foregoing embodiment, which is not described in detail in this embodiment.
The commodity interestingness prediction method of the embodiment further comprises the step of generating the user relationship network diagram, so that the method of the embodiment has the function of generating the user relationship network diagram, and the application range of the method of the embodiment can be effectively expanded.
As an alternative embodiment, referring to fig. 4, the method for predicting the interest level of a commodity further includes a step of recommending commodity information according to a prediction result of the interest level of the commodity, specifically:
In this embodiment, target users are correspondingly determined according to the commodity interest degrees of the users for the commodities, which are included in the interest degree prediction results, and the target users are objects of recommending commodity information and are interested in the commodities and have purchasing intentions. Specifically, some users having a large value of the commodity interest level in the commodity interest level prediction result may be determined as target users. For the selection of the target user, a commodity interest degree threshold value can be set, and the user with the commodity interest degree greater than the commodity interest degree threshold value is determined as the target user.
In this embodiment, a commodity corresponding to the commodity interest level is determined according to the commodity interest level prediction result, and commodity recommendation information is generated. Specifically, the content of the commodity recommendation information may be recommended purchase, recommended trial, or the like.
And 402, sending the commodity recommendation information to the target user.
In this embodiment, the commodity recommendation information is sent to the target user. The specific sending form can be short message, mail or message in software.
The method for predicting the commodity interest degree further comprises the step of recommending the commodity information according to the commodity interest degree prediction result, so that the method of the embodiment can further realize the function of recommending the commodity information after the commodity interest degree is predicted, and the practicability of the method of the embodiment is improved.
Based on the same inventive concept, with reference to fig. 5, an embodiment of the present specification further provides a target user prediction apparatus, including:
an obtaining module 501 configured to obtain a user relationship network map; the user relationship network graph comprises: the system comprises user nodes representing users and edges connecting the user nodes to represent that friend relationships exist among the users; wherein, part of the user nodes are associated with labels for expressing commodity interestingness;
an influence degree determining module 502 configured to input the user relationship network graph into a graph attention network model to obtain an influence degree between the two user nodes connected with the edge;
a weighting module 503, configured to assign a weight to the edge in the user relationship network graph according to the influence degree, so as to obtain a weighted user relationship network graph;
and the predicting module 504 is configured to predict a commodity interest degree prediction result according to the weighted user relationship network diagram.
As an optional embodiment, the apparatus further comprises: the system comprises a generating module, a judging module and a judging module, wherein the generating module is configured to acquire friend relation data of users, and determine users and friend relations among the users; acquiring historical transaction data and determining part of transaction records of the user; generating the user relationship network graph according to the friend relationship between the user and the user; associating the tag for a portion of the user nodes based on the transaction record.
As an optional embodiment, the weight is an influence between two user nodes connected to the edge.
As an alternative embodiment, the algorithm used for prediction is any one of a label propagation algorithm, a full-link search algorithm, and a community structure discovery algorithm.
As an optional embodiment, the commodity interestingness prediction result includes: commodity interestingness of user nodes not associated with the label.
As an optional embodiment, the apparatus further comprises: the recommendation module is configured to determine a target user according to the commodity interest degree prediction result and generate commodity recommendation information; and sending the commodity recommendation information to the target user.
The apparatus of the foregoing embodiment is used to implement the corresponding method in the foregoing embodiment, and has the beneficial effects of the corresponding method embodiment, which are not described herein again.
Based on the same inventive concept, embodiments of the present specification further provide an electronic device, which includes an electronic device memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the program, the method according to any one of the above embodiments is implemented.
Fig. 6 is a schematic diagram illustrating a more specific hardware structure of an electronic device according to this embodiment, where the electronic device may include: a processor 1010, a memory 1020, an input/output interface 1030, a communication interface 1040, and a bus 1050. Wherein the processor 1010, memory 1020, input/output interface 1030, and communication interface 1040 are communicatively coupled to each other within the device via bus 1050.
The processor 1010 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an Application Specific Integrated Circuit (ASIC), or one or more Integrated circuits, and is configured to execute related programs to implement the technical solutions provided in the embodiments of the present disclosure.
The Memory 1020 may be implemented in the form of a ROM (Read Only Memory), a RAM (Random Access Memory), a static storage device, a dynamic storage device, or the like. The memory 1020 may store an operating system and other application programs, and when the technical solutions provided by the embodiments of the present specification are implemented by software or firmware, the relevant program codes are stored in the memory 1020 and called by the processor 1010 for execution.
The input/output interface 1030 is used for connecting an input/output module to input and output information. The i/o module may be configured as a component in a device (not shown) or may be external to the device to provide a corresponding function. The input devices may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output devices may include a display, a speaker, a vibrator, an indicator light, etc.
The communication interface 1040 is used for connecting a communication module (not shown in the drawings) to implement communication interaction between the present apparatus and other apparatuses. The communication module can realize communication in a wired mode (such as USB, network cable and the like) and also can realize communication in a wireless mode (such as mobile network, WIFI, bluetooth and the like).
It should be noted that although the above-mentioned device only shows the processor 1010, the memory 1020, the input/output interface 1030, the communication interface 1040 and the bus 1050, in a specific implementation, the device may also include other components necessary for normal operation. In addition, those skilled in the art will appreciate that the above-described apparatus may also include only those components necessary to implement the embodiments of the present description, and not necessarily all of the components shown in the figures.
The electronic device of the foregoing embodiment is used to implement the corresponding method in the foregoing embodiment, and has the beneficial effects of the corresponding method embodiment, which are not described herein again.
It should be noted that the method of the embodiment of the present invention may be executed by a single device, such as a computer or a server. The method of the embodiment can also be applied to a distributed scene and completed by the mutual cooperation of a plurality of devices. In the case of such a distributed scenario, one of the multiple devices may only perform one or more steps of the method according to the embodiment of the present invention, and the multiple devices interact with each other to complete the method.
The foregoing description has been directed to specific embodiments of this disclosure. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims can be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
Those of ordinary skill in the art will understand that: the discussion of any embodiment above is meant to be exemplary only, and is not intended to intimate that the scope of the disclosure, including the claims, is limited to these examples; within the idea of the invention, also technical features in the above embodiments or in different embodiments may be combined, steps may be implemented in any order, and there are many other variations of the different aspects of the invention as described above, which are not provided in detail for the sake of brevity.
While the present invention has been described in conjunction with specific embodiments thereof, many alternatives, modifications, and variations of these embodiments will be apparent to those of ordinary skill in the art in light of the foregoing description. For example, other memory architectures, such as Dynamic RAM (DRAM), may use the discussed embodiments.
The embodiments of the invention are intended to embrace all such alternatives, modifications and variances that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, substitutions, improvements and the like that may be made without departing from the spirit and principles of the invention are intended to be included within the scope of the invention.
Claims (13)
1. A commodity interestingness prediction method comprises the following steps:
acquiring a user relationship network diagram; the user relationship network graph comprises: the system comprises user nodes representing users and edges connecting the user nodes to represent that friend relationships exist among the users; wherein, part of the user nodes are associated with labels for expressing commodity interestingness;
inputting the user relationship network graph into a graph attention network model to obtain the influence degree between two user nodes connected with the edges;
according to the influence degree, giving a weight to the edge in the user relationship network graph to obtain a weighted user relationship network graph;
and predicting to obtain a commodity interest degree prediction result according to the weighted user relationship network graph.
2. The commodity interestingness prediction method of claim 1, further comprising:
acquiring friend relation data of users, and determining the users and friend relations among the users;
acquiring historical transaction data and determining part of transaction records of the user;
generating the user relationship network graph according to the friend relationship between the user and the user;
associating the tag for a portion of the user nodes based on the transaction record.
3. The method of predicting commodity interestingness according to claim 1, wherein the weight is an influence degree between two user nodes connected to the edge.
4. The method for predicting interest level of a commodity according to claim 1, wherein an algorithm used for prediction is any one of a label propagation algorithm, a comprehensive connection search algorithm and a community structure discovery algorithm.
5. The commodity interestingness prediction method of claim 1, wherein the commodity interestingness prediction result comprises: commodity interestingness of user nodes not associated with the label.
6. The commodity interestingness prediction method of claim 1, further comprising:
determining a target user according to the commodity interest degree prediction result, and generating commodity recommendation information;
and sending the commodity recommendation information to the target user.
7. A commodity interestingness prediction apparatus comprising:
the obtaining module is configured to obtain a user relationship network diagram; the user relationship network graph comprises: the system comprises user nodes representing users and edges connecting the user nodes to represent that friend relationships exist among the users; wherein, part of the user nodes are associated with labels for expressing commodity interestingness;
the influence degree determining module is configured to input the user relationship network graph into a graph attention network model to obtain the influence degree between the two user nodes connected with the edges;
the weighting module is configured to endow the edges in the user relationship network graph with weights according to the influence degree to obtain a weighted user relationship network graph;
and the prediction module is configured to predict a commodity interest degree prediction result according to the weighted user relationship network diagram.
8. The apparatus of claim 7, further comprising:
the system comprises a generating module, a judging module and a judging module, wherein the generating module is configured to acquire friend relation data of users, and determine users and friend relations among the users; acquiring historical transaction data and determining part of transaction records of the user; generating the user relationship network graph according to the friend relationship between the user and the user; associating the tag for a portion of the user nodes based on the transaction record.
9. The apparatus of claim 7, wherein the weight is an influence between two connected user nodes of the edge.
10. The apparatus of claim 7, wherein the algorithm used for prediction is any one of a label propagation algorithm, a full-scale connection search algorithm, and a community structure discovery algorithm.
11. The apparatus of claim 7, the commodity interestingness prediction comprising: commodity interestingness of user nodes not associated with the label.
12. The apparatus of claim 7, further comprising:
the recommendation module is configured to determine a target user according to the commodity interest degree prediction result and generate commodity recommendation information; and sending the commodity recommendation information to the target user.
13. An electronic device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, the processor implementing the method of any one of claims 1 to 6 when executing the program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911375799.1A CN111026973B (en) | 2019-12-27 | 2019-12-27 | Commodity interest degree prediction method and device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911375799.1A CN111026973B (en) | 2019-12-27 | 2019-12-27 | Commodity interest degree prediction method and device and electronic equipment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111026973A CN111026973A (en) | 2020-04-17 |
| CN111026973B true CN111026973B (en) | 2023-04-18 |
Family
ID=70194437
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911375799.1A Active CN111026973B (en) | 2019-12-27 | 2019-12-27 | Commodity interest degree prediction method and device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111026973B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111708876B (en) * | 2020-06-16 | 2023-07-18 | 北京百度网讯科技有限公司 | Method and device for generating information |
| CN112036642A (en) * | 2020-08-31 | 2020-12-04 | 中国平安人寿保险股份有限公司 | Information prediction method, device, equipment and medium based on artificial intelligence |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9031951B1 (en) * | 2012-04-02 | 2015-05-12 | Google Inc. | Associating interest and disinterest keywords with similar and dissimilar users |
| WO2016191959A1 (en) * | 2015-05-29 | 2016-12-08 | 深圳市汇游智慧旅游网络有限公司 | Time-varying collaborative filtering recommendation method |
| CN107895038A (en) * | 2017-11-29 | 2018-04-10 | 四川无声信息技术有限公司 | A kind of link prediction relation recommends method and device |
| CN108198041A (en) * | 2018-01-23 | 2018-06-22 | 重庆裕嘉隆大数据产业有限公司 | A kind of Method of Commodity Recommendation based on time attenuator, terminal and medium |
| CN109977283A (en) * | 2019-03-14 | 2019-07-05 | 中国人民大学 | A kind of the tourism recommended method and system of knowledge based map and user's footprint |
| CN110009474A (en) * | 2019-01-31 | 2019-07-12 | 阿里巴巴集团控股有限公司 | A kind of assessing credit risks method, apparatus and electronic equipment |
| CN110399565A (en) * | 2019-07-29 | 2019-11-01 | 北京理工大学 | Based on when null cycle attention mechanism recurrent neural network point of interest recommended method |
| CN110502701A (en) * | 2019-07-08 | 2019-11-26 | 华南师范大学 | Introduce friend recommendation method, system and the storage medium of attention mechanism |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10848590B2 (en) * | 2005-10-26 | 2020-11-24 | Cortica Ltd | System and method for determining a contextual insight and providing recommendations based thereon |
| US8880532B2 (en) * | 2011-06-29 | 2014-11-04 | International Business Machines Corporation | Interestingness of data |
| US20170039278A1 (en) * | 2012-12-26 | 2017-02-09 | Google Inc. | Annotating social graph edges with interest graph information |
| CN105474213B (en) * | 2013-07-30 | 2019-12-03 | 柯达阿拉里斯股份有限公司 | For creating the system and method that can manipulate view |
| JP5926755B2 (en) * | 2014-03-19 | 2016-05-25 | 国立大学法人京都大学 | Object display system for relationship graph |
| US20160132908A1 (en) * | 2014-11-11 | 2016-05-12 | Mastercard International Incorporated | Methods And Apparatus For Transaction Prediction |
| US10248910B2 (en) * | 2015-10-28 | 2019-04-02 | Fractal Industries, Inc. | Detection mitigation and remediation of cyberattacks employing an advanced cyber-decision platform |
-
2019
- 2019-12-27 CN CN201911375799.1A patent/CN111026973B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9031951B1 (en) * | 2012-04-02 | 2015-05-12 | Google Inc. | Associating interest and disinterest keywords with similar and dissimilar users |
| WO2016191959A1 (en) * | 2015-05-29 | 2016-12-08 | 深圳市汇游智慧旅游网络有限公司 | Time-varying collaborative filtering recommendation method |
| CN107895038A (en) * | 2017-11-29 | 2018-04-10 | 四川无声信息技术有限公司 | A kind of link prediction relation recommends method and device |
| CN108198041A (en) * | 2018-01-23 | 2018-06-22 | 重庆裕嘉隆大数据产业有限公司 | A kind of Method of Commodity Recommendation based on time attenuator, terminal and medium |
| CN110009474A (en) * | 2019-01-31 | 2019-07-12 | 阿里巴巴集团控股有限公司 | A kind of assessing credit risks method, apparatus and electronic equipment |
| CN109977283A (en) * | 2019-03-14 | 2019-07-05 | 中国人民大学 | A kind of the tourism recommended method and system of knowledge based map and user's footprint |
| CN110502701A (en) * | 2019-07-08 | 2019-11-26 | 华南师范大学 | Introduce friend recommendation method, system and the storage medium of attention mechanism |
| CN110399565A (en) * | 2019-07-29 | 2019-11-01 | 北京理工大学 | Based on when null cycle attention mechanism recurrent neural network point of interest recommended method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111026973A (en) | 2020-04-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| TWI718422B (en) | Method, device and equipment for fusing model prediction values | |
| KR20200123015A (en) | Information recommendation method, apparatus, device and medium | |
| CN110008973B (en) | Model training method, method and device for determining target user based on model | |
| US20240095490A1 (en) | Aspect Pre-selection using Machine Learning | |
| CN111026973B (en) | Commodity interest degree prediction method and device and electronic equipment | |
| CN109934646B (en) | Method and device for predicting associated purchasing behavior of new commodity | |
| CA3135466A1 (en) | User loan willingness prediction method and device and computer system | |
| CN111292168B (en) | Data processing method, device and equipment | |
| CN115422334A (en) | Information processing method, device, electronic equipment and storage medium | |
| CN114417174B (en) | Content recommendation method, device, equipment and computer storage medium | |
| CN106600360B (en) | Method and device for sorting recommended objects | |
| KR20210144330A (en) | Method and apparatus for recommending item based on collaborative filtering neural network | |
| CN111461757A (en) | Information processing method and device, computer storage medium and electronic equipment | |
| CN111369258A (en) | Entity object type prediction method, device and equipment | |
| CN113781149A (en) | Information recommendation method and device, computer-readable storage medium and electronic equipment | |
| CN111967924A (en) | Commodity recommendation method, commodity recommendation device, computer device, and medium | |
| CN114429384A (en) | Intelligent product recommendation method and system based on e-commerce platform | |
| CN111199421B (en) | Social relationship-based user recommendation method and device and electronic equipment | |
| CN113792952A (en) | Method and apparatus for generating a model | |
| CN110782287A (en) | Entity similarity calculation method and device, article recommendation system, medium and equipment | |
| CN111325614A (en) | Recommendation method and device of electronic object and electronic equipment | |
| CN111339432A (en) | Recommendation method and device of electronic object and electronic equipment | |
| CN111915339A (en) | Data processing method, device and equipment | |
| CN111159558A (en) | Recommendation list generation method and device and electronic equipment | |
| CN115907892A (en) | User intention identification method and device based on user multi-interest evolution |
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 |