CN113240490A - Intelligent service transaction recommendation method and system based on graph neural network - Google Patents

Intelligent service transaction recommendation method and system based on graph neural network Download PDF

Info

Publication number
CN113240490A
CN113240490A CN202110539216.5A CN202110539216A CN113240490A CN 113240490 A CN113240490 A CN 113240490A CN 202110539216 A CN202110539216 A CN 202110539216A CN 113240490 A CN113240490 A CN 113240490A
Authority
CN
China
Prior art keywords
user
recommendation
model
neural network
representing
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
Application number
CN202110539216.5A
Other languages
Chinese (zh)
Inventor
秦波
张静
王李笑阳
张千一
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Renmin University of China
Original Assignee
Renmin University of China
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Renmin University of China filed Critical Renmin University of China
Priority to CN202110539216.5A priority Critical patent/CN113240490A/en
Publication of CN113240490A publication Critical patent/CN113240490A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00—Commerce
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0631—Item recommendations
    • 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
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • 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
    • G06Q30/02—Marketing; Price estimation or determination; Fundraising
    • G06Q30/0201—Market modelling; Market analysis; Collecting market data
    • G06Q30/0202—Market predictions or forecasting for commercial activities

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • Development Economics (AREA)
  • Strategic Management (AREA)
  • Data Mining & Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Biomedical Technology (AREA)
  • Health & Medical Sciences (AREA)
  • Computational Linguistics (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Biophysics (AREA)
  • Molecular Biology (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computing Systems (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • General Health & Medical Sciences (AREA)
  • General Business, Economics & Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Evolutionary Computation (AREA)
  • Game Theory and Decision Science (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention relates to an intelligent service transaction recommendation method and system based on a graph neural network, which comprises the following steps: preprocessing service transaction data to obtain commodity purchasing information of a user; initializing a graph neural network model according to commodity purchasing information of a user to obtain a score of whether the user purchases the commodity; training the graph neural network model by using the initial embedded vector to obtain an optimal recommendation model; and inputting the user transaction data into the optimal recommendation model for prediction to obtain a recommendation result. The method and the device improve the accuracy of recommendation and can better mine some potential requirements of the user. Meanwhile, the invention omits the processes of feature conversion and nonlinear activation, obviously improves the training efficiency and the prediction efficiency compared with other methods, and can be widely applied to the field of intelligent service transaction data recommendation.

Description

Intelligent service transaction recommendation method and system based on graph neural network
Technical Field
The invention relates to the field of data recommendation, in particular to an intelligent service transaction recommendation method and system based on a graph neural network.
Background
In recent years, with the continuous development of electronic commerce technology, the number of network products and consumers has increased explosively, and both consumers and merchants face huge amounts of information. It is difficult for consumers to sort out the appropriate goods among a large number of goods and for merchants to locate targeted customers among a large number of users. Therefore, an intelligent service transaction recommendation method is needed to find potential relations in transaction data, realize deep mining of user differentiation requirements and accurate positioning of service products, and help merchants and potential consumers to complete transactions. Modern internet service providers provide a large number of commodities for users to select, if the commodities are connected with all consumers who have bought the commodities, a graph is obtained, and higher recommendation accuracy can be obtained by recommending through graph neural network learning.
The current intelligent service transaction recommendation method mainly has the following problems:
1) in some intelligent service transaction recommendation methods, only the purchase records of the users are considered, and users with similar purchase records are not considered at the same time, so that part of potential requirements of the users are not mined.
2) Some intelligent service transaction recommendation methods are complicated in method, and some steps (such as feature transformation and nonlinear activation) have little influence on the collaborative filtering performance, but increase the training difficulty.
Disclosure of Invention
In view of the above problems, an object of the present invention is to provide an intelligent service transaction recommendation method and system based on a graph neural network, which can improve the accuracy of recommendation and further can mine some potential requirements of users.
In order to achieve the purpose, the invention adopts the following technical scheme: a graph neural network-based intelligent service transaction recommendation method comprises the following steps: step 1, preprocessing service transaction data to obtain commodity purchasing information of a user; step 2, initializing a graph neural network model according to the information of the commodities purchased by the user, and obtaining a score of whether the user will purchase the commodities; step 3, training the graph neural network model by using the initial ID embedded vector generated in the step 2 to obtain an optimal recommendation model; and 4, inputting the user transaction data into the optimal recommendation model for prediction to obtain a recommendation result.
Further, in step 1, the data preprocessing method includes the following steps:
step 1.1, respectively endowing different IDs of a user and a commodity in service transaction data;
step 1.2, generating ID pair according to purchase information of user<Ui,Ij>(ii) a Wherein, UiID, I, representing the ith userjAn ID representing the jth product is identified,<Ui,Ij>representing user UiPurchases goods Ij。
Further, in step 2, the method for initializing the neural network model of the graph includes the following steps:
step 2.1, generating an initial ID embedding vector according to the ID pair;
2.2, constructing a lightweight graph convolution layer according to the initial ID embedded vector;
step 2.3, combining the user embedded vectors of all layers into a final layer aggregation vector in layer aggregation;
step 2.4, determining a scoring formula according to the final layer aggregation vector;
score y of whether user u will purchase item iu,iUsing the embedded vector inner product representation:
Figure BDA0003070973250000021
in the formula, yu,iThe expressing model scores the desire of the user u to purchase the commodity i;
Figure BDA0003070973250000022
denotes euThe transposing of (1).
Further, in step 2.2, the method for constructing the lightweight graph convolution layer according to the initial ID embedded vector includes:
Figure BDA0003070973250000023
Figure BDA0003070973250000024
wherein the content of the first and second substances,
Figure BDA0003070973250000025
representing user-embedded vectors after passing through a k-layer network model, in which
Figure BDA0003070973250000026
Representing commodity embedded vectors, N, after passing through a k-layer network modeluRepresenting a set of items purchased by user u, NiRepresenting a set of users who purchased item i.
Further, in step 2.3, the combination formula adopted is:
Figure BDA0003070973250000027
Figure BDA0003070973250000028
wherein alpha iskRepresents the weight of the k layer and represents the weight of the k layer,
Figure BDA0003070973250000029
representing user-embedded vectors after passing through a k-layer network model, in which
Figure BDA00030709732500000210
Representing the commodity embedding vector after passing through the k-layer network model, euRepresenting the aggregated user-embedded vector, eiRepresenting the aggregated commodity embedded vector, K representing the layer set by the network modelAnd (4) counting.
Further, in the step 3, the method for training the neural network model of the graph includes the following steps:
step 3.1, constructing a loss function;
and 3.2, training the selected graph neural network model by using the initial ID embedded vector generated in the step 2.1, and adjusting initial parameters of the machine learning model according to a training result to obtain an optimal recommendation model.
Further, in the step 4, the method for predicting or classifying by using the optimal recommendation model according to the user transaction data includes: inputting data of commodities purchased by a user into a preferred recommendation model, wherein the model feeds back the purchasing desire score of the user for each commodity; if the user's purchasing behavior has been used for training, the scores are fed back directly.
Further, the method comprises the steps of selecting a loss function; and (3) learning and training the graph neural network model by adopting a Bayesian personalized ranking loss function and the summarized data, then participating in the transaction process of certain commodities by the user, and obtaining the commodities which the user may be interested in according to the transaction data and the graph neural network model.
Further, the personalized ranking loss function L of BayesBPRComprises the following steps:
Figure BDA0003070973250000031
where λ represents the regularization strength of the loss function, σ represents the nonlinear activation function, yujA score indicating whether user u will purchase item j.
An intelligent service transaction recommendation system based on a graph neural network is used for realizing the recommendation method, and comprises a user side and a server side; the user side is provided with an individualized recommendation system which is used for displaying suitable commodities for the user and recording operations such as user purchase and the like according to the recommendation ranking and uploading data to the server side; the server is provided with a model training and predicting service system which is used for obtaining an optimal recommendation model according to an existing data training model, predicting or classifying user data by adopting the optimal recommendation model, and returning a prediction or classification result to the user side.
Due to the adoption of the technical scheme, the invention has the following advantages:
1. the recommendation system based on the graph neural network is adopted to recommend the users by mainly considering the behavior of purchasing commodities of similar users, so that the recommendation accuracy is improved, and some potential requirements of the users can be better mined.
2. The invention omits the processes of feature conversion and nonlinear activation, obviously improves the training efficiency and the prediction efficiency compared with other methods in the prior art, and can carry out training and prediction more quickly and better.
In conclusion, the method and the device can be widely applied to the field of intelligent service transaction data recommendation.
Drawings
FIG. 1 is a schematic flow chart of a method for recommending intelligent service transactions in an embodiment of the present invention;
FIG. 2 is a schematic diagram of a model training process in an embodiment of the invention;
FIG. 3 is a flow chart of data encryption and predictive classification in an embodiment of the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the drawings of the embodiments of the present invention. It is to be understood that the embodiments described are only a few embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the described embodiments of the invention, are within the scope of the invention.
The recommendation system based on the graph neural network is adopted to take the behaviors of similar users for purchasing commodities into consideration for recommending the users, so that the recommendation accuracy is improved, and some potential requirements of the users can be better mined. In addition, the invention omits the characteristic transformation and the nonlinear activation process, and the training efficiency and the prediction efficiency are obviously improved compared with other methods.
In a first embodiment of the present invention, as shown in fig. 1, there is provided a graph neural network-based intelligent service transaction recommendation method, which includes the following steps:
step 1, preprocessing service transaction data to obtain commodity purchasing information of a user;
step 2, obtaining an initial ID embedded vector according to the information of the commodity purchased by the user, and initializing a graph neural network model to obtain a score of whether the user will purchase the commodity;
step 3, training the graph neural network model by using the initial ID embedded vector generated in the step 2 to obtain an optimal recommendation model;
and 4, inputting the user transaction data into the optimal recommendation model for prediction to obtain a recommendation result.
In the step 1, the data preprocessing method includes the following steps:
step 1.1, respectively endowing different IDs of a user and a commodity in service transaction data;
step 1.2, generating ID pair according to purchase information of user<Ui,Ij>(ii) a Wherein, UiID, I, representing the ith userjAn ID representing the jth product is identified,<Ui,Ij>representing user UiPurchases goods Ij。
In the step 2, the initialization method for obtaining the initial ID embedded vector and the graph neural network model comprises the following steps:
step 2.1, generating an initial ID embedding vector according to the ID pair;
if the number of users in the service transaction data is M, the number of commodities is N, and the required embedding size is H, then the matrix E is randomly selected(0)∈R(M+N)×H. At initialization matrix E(0)In, initial user embedded vector usage
Figure BDA0003070973250000041
Representing, initial commodity embedding vector usage
Figure BDA0003070973250000042
And (4) showing. R represents a real number set.
2.2, constructing a lightweight graph convolution layer according to the initial ID embedded vector;
the construction method is shown in the following formula:
Figure BDA0003070973250000043
Figure BDA0003070973250000044
wherein the content of the first and second substances,
Figure BDA0003070973250000045
representing user-embedded vectors after passing through a k-layer network model, in which
Figure BDA0003070973250000046
Representing commodity embedded vectors, N, after passing through a k-layer network modeluRepresenting a set of items purchased by user u, NiRepresenting a set of users who purchased item i.
Step 2.3, combining the user embedded vectors of all layers into a final layer aggregation vector in layer aggregation;
the combination formula adopted is:
Figure BDA0003070973250000047
Figure BDA0003070973250000048
wherein alpha iskRepresenting the weight of k layers, euRepresenting the aggregated user-embedded vector, eiAnd K represents the number of layers set by the network model.
Step 2.4, determining a scoring formula according to the final layer aggregation vector;
the score of whether a user u will purchase an item i is expressed using the embedded vector inner product, namely:
Figure BDA0003070973250000051
in the formula, yu,iThe expressing model scores the desire of the user u to purchase the commodity i;
Figure BDA0003070973250000052
denotes euThe transposing of (1).
In the step 3, as shown in fig. 2, the method for training the neural network model includes the following steps:
step 3.1, constructing a loss function;
and 3.2, training the graph neural network model initialized in the step 2 by using the initial ID embedded vector generated in the step 2.1, and adjusting initial parameters of the machine learning model according to a training result to achieve an optimal effect and obtain an optimal recommendation model.
In the step 4, the method for predicting or classifying by using the optimal recommendation model according to the user transaction data comprises the following steps: and inputting data of commodities purchased by the user into a preferred recommendation model, wherein the model feeds back the purchasing desire score of the user for each commodity. If the user's purchasing behavior has been used for training, the scores are fed back directly.
In a preferred embodiment, as shown in fig. 3, a step of selecting a loss function is further included. In the embodiment, a Bayesian personalized ranking loss function is adopted, the summarized data is adopted to carry out learning training of the graph neural network model, then the user participates in the transaction process of certain commodities, and the commodities which the user may be interested in are obtained according to the transaction data and the graph neural network model.
In the embodiment, the transaction record of the commodity purchased by the user is used as a data source, and the personalized ranking loss function L of Bayesian is adoptedBPRComprises the following steps:
Figure BDA0003070973250000053
where λ represents the regularization strength of the loss function, σ represents the nonlinear activation function, yujA score indicating whether user u will purchase item j.
In a second embodiment of the present invention, an intelligent service transaction recommendation system based on a graph neural network is provided, which is used for implementing the recommendation method in the first embodiment. The recommendation system comprises a user side and a server side; the user side is provided with an individualized recommendation system which is used for displaying suitable commodities for the user and recording operations such as user purchase and the like according to the recommendation ranking and uploading data to the server side; the server is provided with a model training and predicting service system which is used for obtaining an optimal recommendation model according to the existing data training model, predicting or classifying the user data by adopting the optimal recommendation model and returning the prediction or classification result to the user side.
In the above embodiment, the personalized recommendation system includes a user data uploading module and a personalized goods display module. The user data uploading module is used for recording clicking or purchasing operation of the user on the commodity and uploading the record to the server; and the personalized commodity display is used for displaying commodities which are more likely to be purchased by the user for the user according to the user purchasing desire score returned by the server.
In the above embodiment, the model training and prediction service system includes a model training module and a prediction scoring module. The model training module is used for training the recommendation model according to the summarized data of user purchase or click and the like to obtain an optimal transaction recommendation model; and the prediction scoring module is used for predicting or scoring the user purchase data according to the optimal recommendation model to obtain a prediction or scoring result, and the scoring result is returned to the client.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

Claims (10)

1. An intelligent service transaction recommendation method based on a graph neural network is characterized by comprising the following steps:
step 1, preprocessing service transaction data to obtain commodity purchasing information of a user;
step 2, initializing a graph neural network model according to the information of the commodities purchased by the user, and obtaining a score of whether the user will purchase the commodities;
step 3, training the graph neural network model by using the initial ID embedded vector generated in the step 2 to obtain an optimal recommendation model;
and 4, inputting the user transaction data into the optimal recommendation model for prediction to obtain a recommendation result.
2. The recommendation method as claimed in claim 1, wherein in the step 1, the data preprocessing method comprises the steps of:
step 1.1, respectively endowing different IDs of a user and a commodity in service transaction data;
step 1.2, generating ID pair according to purchase information of user<Ui,Ij>(ii) a Wherein, UiID, I, representing the ith userjAn ID representing the jth product is identified,<Ui,Ij>representing user UiPurchases goods Ij。
3. The recommendation method according to claim 1, wherein in the step 2, the neural network model initialization method comprises the following steps:
step 2.1, generating an initial ID embedding vector according to the ID pair;
2.2, constructing a lightweight graph convolution layer according to the initial ID embedded vector;
step 2.3, combining the user embedded vectors of all layers into a final layer aggregation vector in layer aggregation;
step 2.4, determining a scoring formula according to the final layer aggregation vector;
score y of whether user u will purchase item iu,iUsing the embedded vector inner product representation:
Figure FDA0003070973240000011
in the formula, yu,iThe expressing model scores the desire of the user u to purchase the commodity i;
Figure FDA0003070973240000012
denotes euThe transposing of (1).
4. The recommendation method according to claim 3, wherein in step 2.2, the method for constructing the lightweight map convolutional layer according to the initial ID embedded vector comprises:
Figure FDA0003070973240000013
Figure FDA0003070973240000014
wherein the content of the first and second substances,
Figure FDA0003070973240000015
representing user-embedded vectors after passing through a k-layer network model, in which
Figure FDA0003070973240000016
Representing commodity embedded vectors, N, after passing through a k-layer network modeluRepresenting a set of items purchased by user u, NiRepresenting a set of users who purchased item i.
5. A recommendation method according to claim 3, characterised in that in step 2.3, the combined formula used is:
Figure FDA0003070973240000021
Figure FDA0003070973240000022
wherein alpha iskRepresents the weight of the k layer and represents the weight of the k layer,
Figure FDA0003070973240000023
representing user-embedded vectors after passing through a k-layer network model, in which
Figure FDA0003070973240000024
Representing the commodity embedding vector after passing through the k-layer network model, euRepresenting the aggregated user-embedded vector, eiAnd K represents the number of layers set by the network model.
6. The recommendation method according to claim 1, wherein in the step 3, the method for training the neural network model comprises the following steps:
step 3.1, constructing a loss function;
and 3.2, training the selected graph neural network model by using the initial ID embedded vector generated in the step 2.1, and adjusting initial parameters of the machine learning model according to a training result to obtain an optimal recommendation model.
7. The recommendation method of claim 1, wherein in step 4, the method for predicting or classifying according to the user transaction data by using the optimal recommendation model comprises: inputting data of commodities purchased by a user into a preferred recommendation model, wherein the model feeds back the purchasing desire score of the user for each commodity; if the user's purchasing behavior has been used for training, the scores are fed back directly.
8. The recommendation method of claim 1, further comprising the step of selecting a loss function; and (3) learning and training the graph neural network model by adopting a Bayesian personalized ranking loss function and the summarized data, then participating in the transaction process of certain commodities by the user, and obtaining the commodities which the user may be interested in according to the transaction data and the graph neural network model.
9. The recommendation method of claim 8, wherein the bayesian personalized ranking loss function LBPRComprises the following steps:
Figure FDA0003070973240000025
where λ represents the regularization strength of the loss function, σ represents the nonlinear activation function, yujA score indicating whether user u will purchase item j.
10. An intelligent service transaction recommendation system based on a graph neural network, which is used for implementing the recommendation method according to any one of claims 1 to 9, and comprises a user side and a service side; the user side is provided with an individualized recommendation system which is used for displaying suitable commodities for the user and recording operations such as user purchase and the like according to the recommendation ranking and uploading data to the server side; the server is provided with a model training and predicting service system which is used for obtaining an optimal recommendation model according to an existing data training model, predicting or classifying user data by adopting the optimal recommendation model, and returning a prediction or classification result to the user side.
CN202110539216.5A 2021-05-18 2021-05-18 Intelligent service transaction recommendation method and system based on graph neural network Pending CN113240490A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110539216.5A CN113240490A (en) 2021-05-18 2021-05-18 Intelligent service transaction recommendation method and system based on graph neural network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110539216.5A CN113240490A (en) 2021-05-18 2021-05-18 Intelligent service transaction recommendation method and system based on graph neural network

Publications (1)

Publication Number Publication Date
CN113240490A true CN113240490A (en) 2021-08-10

Family

ID=77135012

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110539216.5A Pending CN113240490A (en) 2021-05-18 2021-05-18 Intelligent service transaction recommendation method and system based on graph neural network

Country Status (1)

Country Link
CN (1) CN113240490A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114925279A (en) * 2022-06-07 2022-08-19 支付宝(杭州)信息技术有限公司 Recommendation model training method, recommendation method and recommendation device
CN115545467A (en) * 2022-09-30 2022-12-30 广东工业大学 Risk commodity identification model based on graph neural network
CN116108291A (en) * 2023-04-12 2023-05-12 武汉中科通达高新技术股份有限公司 Mobile internet traffic service recommendation method and system
CN117217710A (en) * 2023-10-19 2023-12-12 深圳市金文网络科技有限公司 Intelligent management method and system for virtual commodity and shortcut service

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108717514A (en) * 2018-05-21 2018-10-30 中国人民大学 A kind of data-privacy guard method in machine learning and system
CN111160954A (en) * 2019-12-16 2020-05-15 南京理工大学 Recommendation method facing group object based on graph convolution network model
CN112561644A (en) * 2020-12-16 2021-03-26 深圳市网联安瑞网络科技有限公司 Commodity recommendation method and device based on link prediction and related equipment
CN112800334A (en) * 2021-02-04 2021-05-14 河海大学 Collaborative filtering recommendation method and device based on knowledge graph and deep learning

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108717514A (en) * 2018-05-21 2018-10-30 中国人民大学 A kind of data-privacy guard method in machine learning and system
CN111160954A (en) * 2019-12-16 2020-05-15 南京理工大学 Recommendation method facing group object based on graph convolution network model
CN112561644A (en) * 2020-12-16 2021-03-26 深圳市网联安瑞网络科技有限公司 Commodity recommendation method and device based on link prediction and related equipment
CN112800334A (en) * 2021-02-04 2021-05-14 河海大学 Collaborative filtering recommendation method and device based on knowledge graph and deep learning

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
XIANGNAN HE: ""LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation"", 《PROCEEDINGS OF THE 43RD INTERNATIONAL ACM SIGIR CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION RETRIEVAL》 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114925279A (en) * 2022-06-07 2022-08-19 支付宝(杭州)信息技术有限公司 Recommendation model training method, recommendation method and recommendation device
CN115545467A (en) * 2022-09-30 2022-12-30 广东工业大学 Risk commodity identification model based on graph neural network
CN115545467B (en) * 2022-09-30 2024-01-23 广东工业大学 Risk commodity identification model based on graphic neural network
CN116108291A (en) * 2023-04-12 2023-05-12 武汉中科通达高新技术股份有限公司 Mobile internet traffic service recommendation method and system
CN116108291B (en) * 2023-04-12 2023-08-08 武汉中科通达高新技术股份有限公司 Mobile internet traffic service recommendation method and system
CN117217710A (en) * 2023-10-19 2023-12-12 深圳市金文网络科技有限公司 Intelligent management method and system for virtual commodity and shortcut service

Similar Documents

Publication Publication Date Title
US10783361B2 (en) Predictive analysis of target behaviors utilizing RNN-based user embeddings
CN113240490A (en) Intelligent service transaction recommendation method and system based on graph neural network
US10635952B2 (en) Cognitive analysis and classification of apparel images
Grupac et al. Image processing computational algorithms, sensory data mining techniques, and predictive customer analytics in the metaverse economy
Policarpo et al. Machine learning through the lens of e-commerce initiatives: An up-to-date systematic literature review
Zvarikova et al. Cognitive artificial intelligence algorithms, movement and behavior tracking tools, and customer identification technology in the metaverse commerce
JP2016522523A (en) Method and system for recommending information
CA3031548A1 (en) System and method for analyzing and searching for features associated with objects
US10275819B1 (en) Reducing incompatible pairings
KR102422408B1 (en) Method and apparatus for recommending item based on collaborative filtering neural network
Grupac et al. Virtual navigation and augmented reality shopping tools, immersive and cognitive technologies, and image processing computational and object tracking algorithms in the metaverse commerce
Wang et al. Webpage depth viewability prediction using deep sequential neural networks
Fareed et al. A collaborative filtering recommendation framework utilizing social networks
CN116764236A (en) Game prop recommending method, game prop recommending device, computer equipment and storage medium
Seema et al. Development of fading channel patch based convolutional neural network models for customer churn prediction
US11887134B2 (en) Product performance with location on page analysis
WO2024006867A1 (en) Modeling to generate dynamic electronic representations
Loukili et al. Personalizing product recommendations using collaborative filtering in online retail: a machine learning approach
Luo et al. Click-Through Rate Prediction Models based on Interest Modeling
Wen et al. Pairwise multi-layer nets for learning distributed representation of multi-field categorical data
Deng et al. Similitude attentive relation network for click-through rate prediction
Gehlot et al. Execution of market basket analysis and recommendation systems in physical retail stores to advance sales revenues
CN113379482B (en) Article recommendation method, computing device and storage medium
US20230245183A1 (en) Systems and methods for generating vehicle buyback guarantees
US20220230226A1 (en) Similar item recommendation framework using wide-and-deep-architecture

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