CN110674407A - Hybrid recommendation method based on graph convolution neural network - Google Patents

Hybrid recommendation method based on graph convolution neural network Download PDF

Info

Publication number
CN110674407A
CN110674407A CN201910940872.9A CN201910940872A CN110674407A CN 110674407 A CN110674407 A CN 110674407A CN 201910940872 A CN201910940872 A CN 201910940872A CN 110674407 A CN110674407 A CN 110674407A
Authority
CN
China
Prior art keywords
user
feature vector
nodes
node
article
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN201910940872.9A
Other languages
Chinese (zh)
Other versions
CN110674407B (en
Inventor
王东京
张新
俞东进
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.)
Hangzhou Dianzi University
Hangzhou Electronic Science and Technology University
Original Assignee
Hangzhou Electronic Science and Technology University
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 Hangzhou Electronic Science and Technology University filed Critical Hangzhou Electronic Science and Technology University
Priority to CN201910940872.9A priority Critical patent/CN110674407B/en
Publication of CN110674407A publication Critical patent/CN110674407A/en
Application granted granted Critical
Publication of CN110674407B publication Critical patent/CN110674407B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • 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
    • G06Q30/06—Buying, selling or leasing transactions
    • G06Q30/0601—Electronic shopping [e-shopping]
    • G06Q30/0631—Item recommendations

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Databases & Information Systems (AREA)
  • Business, Economics & Management (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Finance (AREA)
  • Accounting & Taxation (AREA)
  • Data Mining & Analysis (AREA)
  • Computational Linguistics (AREA)
  • Artificial Intelligence (AREA)
  • Computing Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Molecular Biology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Marketing (AREA)
  • Strategic Management (AREA)
  • General Business, Economics & Management (AREA)
  • Health & Medical Sciences (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The invention discloses a hybrid recommendation method based on a graph convolution neural network, which comprises the following steps of: collecting behavior data of a user on an article and attribute information of the article; modeling the collected data into a heterogeneous information graph, and learning the feature vector representation of each node by utilizing a graph convolution neural network; and recommending based on the mixture of the user and the item feature vector. The method mainly utilizes various data including behavior data of the user to the article and article attribute information to obtain feature vector representation of the user and the article and implement mixed recommendation based on the feature vectors, so that the influence of data sparseness and heavy feature engineering problems is relieved, the recommendation effect is improved, and the user satisfaction is improved.

Description

Hybrid recommendation method based on graph convolution neural network
Technical Field
The invention belongs to the technical field of data mining, information retrieval and recommendation, and particularly relates to a hybrid recommendation method based on a graph convolution neural network.
Background
With the rapid development of information technology, people can enjoy network services and contents conveniently and rapidly, but also face the problem of information overload caused by mass data, and are difficult to find the contents of interest. The recommendation system can help a user to find related data from massive online information to meet the user requirements, and accurately obtaining the characteristics of the articles and efficiently calculating the similarity of the articles is one of the cores for realizing the personalized recommendation system.
However, the conventional method generally suffers from problems such as data sparseness, heavy feature engineering, etc., and cannot meet the user's needs. Therefore, how to fully obtain the multi-source heterogeneous information including the behavior data of the user and the attributes of the articles in an automatic and intelligent manner, accurately obtain the feature vectors of the articles and realize efficient hybrid recommendation is one of the keys of solving the problem of sparse and heavy feature engineering of data, improving the accuracy of a recommendation system and improving the satisfaction degree of the user.
Disclosure of Invention
Aiming at the technical problems in the prior art, the invention provides a hybrid recommendation method based on a graph convolution neural network, which can improve the recommendation effect and performance.
The invention comprises the following contents:
1. a hybrid recommendation method based on a graph convolution neural network comprises the following steps:
10. collecting behavior data of a user and attributes of an article;
20. modeling the collected data and learning feature vector representations of the user and the article by using a graph convolution neural network;
30. and performing mixed recommendation based on the user behavior and the item feature vector.
Wherein the step 10 comprises:
101. collecting behavioral data of all users
Figure BDA0002222856470000021
U＝{u1,u2,...,u|U|Is the set of all users, where the behavior data of user U e U is represented as the sequence Bu＝{(i1,t1),(i2,t2),…,(im,tm)}，(ij,tj) Indicating the item interacted by the user and the interaction time, I ═ I1,i2,...,i|M|Is the collection of all items.
102. Collecting attribute data A of all the items in I, including but not limited to category, label, metadata and other information.
Wherein step 20 comprises:
201. constructing a heterogeneous information graph G (N, E) according to the behavior data of all users and the attribute data of all articles, wherein a node set N (U ∪ I ∪ A) in the heterogeneous information graph G comprises user nodes, article nodes and attribute nodes, and an edge set in the heterogeneous information graphEu,iIs a collection of edges representing the interaction of users with items in the behavioural data,
Figure BDA0002222856470000023
is a set of edges representing the relationship between the items and the item transitions in the sequence of actions, Ei，aIs a collection of edges representing music-attribute dependencies.
202. According to the heterogeneous information graph G, the following objective function O is established:
Figure BDA0002222856470000024
wherein: σ (-) is a sigmoid function,
Figure BDA0002222856470000025
and
Figure BDA0002222856470000026
is node niAnd njOf a d-dimensional graph, in particular, a vector representation v of a node nnIs defined as:
Figure BDA0002222856470000027
wherein: σ (-) is the sigmoid activation function, W is the weight matrix,
Figure BDA0002222856470000028
is a vector-splicing operation that is performed,
Figure BDA0002222856470000031
is the N-dimensional neighboring feature vector for node N,is the mean value of the neighboring feature vectors of neighboring nodes, defined as:
Figure BDA0002222856470000033
wherein:
Figure BDA0002222856470000034
is a set of contiguous nodes to the node n,
Figure BDA0002222856470000035
a N-dimensional neighboring feature vector, w, representing a node Nt(n),t(n′)Representing the degree of interaction between different types of nodes, t (n), t (n ') e (User, Item, Attribute) representing the types of nodes n and n'.
Wherein step 30 comprises:
301. according to the feature vector representation obtained by learning, the interest of the target user u in the item i can be represented by the cosine similarity of the corresponding vector, which is specifically as follows:
p(u,i)＝cos(vu,vi)
wherein the content of the first and second substances,is a d-dimensional graph convolution feature vector representation of user u and item i.
302. And sorting all the articles by using the operation result of the formula, and recommending the first articles to the target user u.
The invention has the beneficial effects that:
1) multisource heterogeneous information including behavior data of a user and attributes of articles is fully utilized, and the problem of data sparsity is solved;
2) the characteristic extraction and recommendation algorithm based on the convolutional neural network is designed, so that heavy characteristic engineering can be relieved, the accuracy and diversity of recommendation results are improved, and the satisfaction degree of users is further improved.
Drawings
FIG. 1 is a schematic diagram of a system architecture of a recommendation method of the present invention.
Detailed Description
In order to more specifically describe the present invention, the following detailed description is provided for the technical solution of the present invention with reference to the accompanying drawings and the specific embodiments.
The mixed recommendation method based on the graph convolution neural network comprises the following steps:
(1) collect behavioral data of all users
Figure BDA0002222856470000041
U＝{u1,u2,...,u|U|The set of all users is denoted with the behavioral data of user U e U as the sequence Bu { (i)1,t1),(i2,t2),…,(im,tm)}，(ij,tj) Respectively representing the item interacted by the user and the interaction time, I ═ I1,i2,...,i|M|Is the collection of all items. Collecting attribute data A of all the items in I, including but not limited to category, label, metadata and other information.
(2) Constructing a heterogeneous information graph G (N, E) according to the behavior data of all users and the attribute data of all articles, wherein a node set N (U ∪ I ∪ A in the heterogeneous information graph G comprises user nodes, article nodes and attribute nodes, and an edge set in the heterogeneous information graph
Figure BDA0002222856470000042
Eu,iIs a collection of edges representing the interaction of users with items in the behavioural data,is to indicate the item and the item transfer in the action sequenceSet of edges of the series, Ei,aIs a collection of edges representing music-attribute dependencies.
(3) The vector for node n may be defined based on a graph convolution neural network as:
Figure BDA0002222856470000044
where W is a matrix of weights,is a vector-splicing operation that is performed,
Figure BDA0002222856470000046
is the N-dimensional neighboring feature vector for node N,
Figure BDA0002222856470000047
is the mean value of the neighboring feature vectors of the neighboring nodes, defined as
Figure BDA0002222856470000048
Wherein
Figure BDA0002222856470000049
Is a set of contiguous nodes to the node n,
Figure BDA00022228564700000410
a N-dimensional neighboring feature vector, w, representing a node Nt(n),t(n′)Representing the degree of interaction between different types of nodes, t (n), t (n ') e (User, Item, Attribute) representing the types of nodes n and n'.
(4) Establishing an objective function according to the graph convolution eigenvector calculation formula of the heterogeneous information graph and the nodes
Figure BDA00022228564700000411
Where σ (-) is a sigmoid function,
Figure BDA0002222856470000051
and
Figure BDA0002222856470000052
is section (III)Point niAnd njThe d-dimensional graph convolution feature vector representation of (1); obtaining a feature vector representation of each node by minimizing the objective function
Figure BDA0002222856470000053
(5) According to the learned feature vector representation, the interest of the target user u in the item i can be represented as p (u, i) ═ cos (v) by the cosine similarity of the corresponding feature vectoru,vi) Wherein
Figure BDA0002222856470000054
Is a feature vector representation of user u and item i.
(6) Sorting all the articles by using the operation result of the formula, and recommending the first articles to the target user u.
Fig. 1 shows an architecture of the hybrid recommendation method based on the graph convolution neural network according to the present embodiment. The recommendation system is divided into two main modules: the device comprises a preprocessing module and a recommending module. In the preprocessing module, firstly behavior data of all users and attribute information of articles are obtained, a heterogeneous information graph is constructed to model various information, and then feature vector representations of user nodes and article nodes in the heterogeneous information graph are obtained by utilizing a graph convolution neural network. In a recommending module, the interest of a prediction target user on candidate articles is represented and ranked based on the node feature vectors obtained through learning, and finally a plurality of articles ranked at the top are recommended to the user.
The embodiments described above are intended to facilitate one of ordinary skill in the art in understanding and using the invention. It will be readily apparent to those skilled in the art that various modifications to the above-described implementations may be made, and the generic principles described herein may be applied to other implementations without the use of inventive faculty. Therefore, the present invention is not limited to the above embodiments, and those skilled in the art should make improvements and modifications to the present invention based on the disclosure of the present invention within the protection scope of the present invention.

Claims (6)

1. The hybrid recommendation method based on the graph convolution neural network is characterized by comprising the following steps of:
10. collecting behavior data of a user and attributes of an article;
20. modeling the collected data and learning feature vector representations of the user and the article by using a graph convolution neural network;
30. and performing mixed recommendation based on the user behavior and the item feature vector.
2. The method of claim 1, wherein step 10 comprises:
101. collecting behavioral data of all usersU＝{u1,u2,...,u|U|Is the set of all users, where the behavior data of user U e U is represented as the sequence Bu＝{(i1,t1),(i2,t2),…,(im,tm)}，(ij,tj) Indicating the item interacted by the user and the interaction time, I ═ I1,i2,...,i|M|Is the collection of all items;
102. collecting attribute data A of all items in I, including but not limited to category, label, metadata.
3. The method of claim 1, wherein step 20 comprises:
201. constructing a heterogeneous information graph G (N, E) according to the behavior data of all users and the attribute data of all articles, wherein a node set N (U ∪ I ∪ A) in the heterogeneous information graph G comprises user nodes, article nodes and attribute nodes, and an edge set in the heterogeneous information graph
Figure FDA0002222856460000012
Eu,iIs to represent users and articles in the behavior dataThe set of edges of the interaction relationship(s),
Figure FDA0002222856460000013
is a set of edges representing the relationship between the items and the item transitions in the sequence of actions, Ei,aIs a set of edges representing music-attribute dependencies;
202. according to the heterogeneous information graph G, the following objective function O is established:
Figure FDA0002222856460000021
wherein: σ (-) is a sigmoid function,and
Figure FDA0002222856460000023
is node niAnd njThe d-dimensional graph convolution feature vector representation of (a) is obtained by minimizing the objective function O to obtain the feature vector representation of each node.
4. The method of claim 3, wherein the vector representation v of the node n in step 20 isnIs defined as:
Figure FDA0002222856460000024
wherein: w is a matrix of weights that is,
Figure FDA0002222856460000025
is a vector-splicing operation that is performed,
Figure FDA0002222856460000026
is the N-dimensional neighboring feature vector for node N,is the neighbor feature vector mean of the neighbor nodes.
5. The hybrid recommendation method based on the graph convolution neural network of claim 4, wherein the neighboring feature vector mean of the neighboring nodes is defined as:
Figure FDA0002222856460000028
wherein:
Figure FDA0002222856460000029
is a set of contiguous nodes to the node n,
Figure FDA00022228564600000210
a N-dimensional neighboring feature vector, w, representing a node Nt(n),t(n′)Representing the degree of interaction between different types of nodes, t (n), t (n ') e (User, Item, Attribute) representing the types of nodes n and n'.
6. The method of claim 1, wherein step 30 comprises:
301. according to the feature vector representation obtained by learning, the interest of the target user u in the item i is represented by the cosine similarity of the corresponding vector, which specifically comprises the following steps:
p(u,i)＝cos(vu,vi)
wherein v isu,
Figure FDA0002222856460000031
Is a d-dimensional graph convolution feature vector representation of a user u and an article i;
302. and sorting all the articles by using the operation result of the formula, and recommending the first articles to the target user u.
CN201910940872.9A 2019-09-30 2019-09-30 Hybrid recommendation method based on graph convolution neural network Active CN110674407B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910940872.9A CN110674407B (en) 2019-09-30 2019-09-30 Hybrid recommendation method based on graph convolution neural network

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910940872.9A CN110674407B (en) 2019-09-30 2019-09-30 Hybrid recommendation method based on graph convolution neural network

Publications (2)

Publication Number Publication Date
CN110674407A true CN110674407A (en) 2020-01-10
CN110674407B CN110674407B (en) 2021-10-15

Family

ID=69078761

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910940872.9A Active CN110674407B (en) 2019-09-30 2019-09-30 Hybrid recommendation method based on graph convolution neural network

Country Status (1)

Country Link
CN (1) CN110674407B (en)

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111291270A (en) * 2020-03-02 2020-06-16 合肥工业大学 Attribute reasoning and product recommendation method based on self-adaptive graph convolution network
CN111611472A (en) * 2020-03-31 2020-09-01 清华大学 Binding recommendation method and system based on graph convolution neural network
CN111737569A (en) * 2020-06-04 2020-10-02 山东省人工智能研究院 Personalized recommendation method based on attribute perception intention-minded convolutional neural network
CN111931076A (en) * 2020-09-22 2020-11-13 平安国际智慧城市科技股份有限公司 Method and device for carrying out relationship recommendation based on authorized directed graph and computer equipment
CN111949865A (en) * 2020-08-10 2020-11-17 杭州电子科技大学 Interest point recommendation method based on graph neural network and user long-term and short-term preference
CN112329928A (en) * 2020-12-30 2021-02-05 四川新网银行股份有限公司 Heterogeneous model-based user satisfaction analysis method
CN112711708A (en) * 2020-12-30 2021-04-27 济南浪潮高新科技投资发展有限公司 Product scheme combination recommendation method based on deep learning
CN113095870A (en) * 2021-03-16 2021-07-09 支付宝(杭州)信息技术有限公司 Prediction method, prediction device, computer equipment and storage medium
CN113297490A (en) * 2021-06-04 2021-08-24 西南大学 Bidirectional recommendation method based on graph convolution neural network
WO2021204039A1 (en) * 2020-04-10 2021-10-14 北京沃东天骏信息技术有限公司 Method and apparatus for pushing information
CN114417161A (en) * 2022-01-21 2022-04-29 杭州碧游信息技术有限公司 Virtual article time sequence recommendation method, device, medium and equipment based on special-purpose map
CN115187343A (en) * 2022-07-20 2022-10-14 山东省人工智能研究院 Multi-behavior recommendation method based on attention map convolution neural network
US11681737B2 (en) 2020-04-08 2023-06-20 Adobe Inc. Electronic media retrieval

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108874914A (en) * 2018-05-29 2018-11-23 吉林大学 A kind of information recommendation method based on the long-pending and neural collaborative filtering of picture scroll
CN109299373A (en) * 2018-10-20 2019-02-01 上海交通大学 Recommender system based on figure convolution technique
CN109446413A (en) * 2018-09-25 2019-03-08 上海交通大学 Serializing recommended method based on item associations relationship

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108874914A (en) * 2018-05-29 2018-11-23 吉林大学 A kind of information recommendation method based on the long-pending and neural collaborative filtering of picture scroll
CN109446413A (en) * 2018-09-25 2019-03-08 上海交通大学 Serializing recommended method based on item associations relationship
CN109299373A (en) * 2018-10-20 2019-02-01 上海交通大学 Recommender system based on figure convolution technique

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
REX YING等: ""Graph Convolutional Neural Networks for Web-Scale Recommender Systems"", 《PROCEEDINGS OF THE 24TH ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY & DATA》 *
江原: ""基于图卷积与神经协同过滤的融合信息推荐模型"", 《中国优秀硕士学位论文全文数据库 信息科技辑》 *

Cited By (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111291270B (en) * 2020-03-02 2021-07-27 合肥工业大学 Attribute reasoning and product recommendation method based on self-adaptive graph convolution network
CN111291270A (en) * 2020-03-02 2020-06-16 合肥工业大学 Attribute reasoning and product recommendation method based on self-adaptive graph convolution network
CN111611472A (en) * 2020-03-31 2020-09-01 清华大学 Binding recommendation method and system based on graph convolution neural network
CN111611472B (en) * 2020-03-31 2024-02-27 清华大学 Binding recommendation method and system based on graph convolution neural network
US11681737B2 (en) 2020-04-08 2023-06-20 Adobe Inc. Electronic media retrieval
WO2021204039A1 (en) * 2020-04-10 2021-10-14 北京沃东天骏信息技术有限公司 Method and apparatus for pushing information
CN111737569A (en) * 2020-06-04 2020-10-02 山东省人工智能研究院 Personalized recommendation method based on attribute perception intention-minded convolutional neural network
CN111737569B (en) * 2020-06-04 2022-05-03 山东省人工智能研究院 Personalized recommendation method based on attribute perception intention-minded convolutional neural network
CN111949865A (en) * 2020-08-10 2020-11-17 杭州电子科技大学 Interest point recommendation method based on graph neural network and user long-term and short-term preference
CN111931076A (en) * 2020-09-22 2020-11-13 平安国际智慧城市科技股份有限公司 Method and device for carrying out relationship recommendation based on authorized directed graph and computer equipment
CN112329928B (en) * 2020-12-30 2021-04-30 四川新网银行股份有限公司 Heterogeneous model-based user satisfaction analysis method
CN112711708A (en) * 2020-12-30 2021-04-27 济南浪潮高新科技投资发展有限公司 Product scheme combination recommendation method based on deep learning
CN112329928A (en) * 2020-12-30 2021-02-05 四川新网银行股份有限公司 Heterogeneous model-based user satisfaction analysis method
CN113095870A (en) * 2021-03-16 2021-07-09 支付宝(杭州)信息技术有限公司 Prediction method, prediction device, computer equipment and storage medium
CN113297490B (en) * 2021-06-04 2022-08-02 西南大学 Bidirectional recommendation method based on graph convolution neural network
CN113297490A (en) * 2021-06-04 2021-08-24 西南大学 Bidirectional recommendation method based on graph convolution neural network
CN114417161A (en) * 2022-01-21 2022-04-29 杭州碧游信息技术有限公司 Virtual article time sequence recommendation method, device, medium and equipment based on special-purpose map
CN115187343A (en) * 2022-07-20 2022-10-14 山东省人工智能研究院 Multi-behavior recommendation method based on attention map convolution neural network
CN115187343B (en) * 2022-07-20 2023-08-08 山东省人工智能研究院 Attention graph convolution neural network-based multi-behavior recommendation method

Also Published As

Publication number Publication date
CN110674407B (en) 2021-10-15

Similar Documents

Publication Publication Date Title
CN110674407B (en) Hybrid recommendation method based on graph convolution neural network
CN110162593B (en) Search result processing and similarity model training method and device
CN103345645B (en) Commodity image class prediction method towards net purchase platform
Zhao et al. Automatic unsupervised outlier model selection
CN105469096B (en) A kind of characteristic bag image search method based on Hash binary-coding
Yin et al. Incomplete multi-view clustering via subspace learning
CN101295305B (en) Image retrieval device
CN110929161B (en) Large-scale user-oriented personalized teaching resource recommendation method
Xiaomei et al. Microblog sentiment analysis with weak dependency connections
Xie et al. A survey on dynamic network embedding
CN106021364A (en) Method and device for establishing picture search correlation prediction model, and picture search method and device
US20110029466A1 (en) Supervised rank aggregation based on rankings
CN104834693A (en) Depth-search-based visual image searching method and system thereof
Ren et al. Heterogeneous graph-based intent learning with queries, web pages and wikipedia concepts
CN112380453B (en) Article recommendation method and device, storage medium and equipment
CN107577786B (en) A kind of matrix decomposition recommendation method based on joint cluster
CN113468227A (en) Information recommendation method, system, device and storage medium based on graph neural network
CN110990670B (en) Growth incentive book recommendation method and recommendation system
WO2023155508A1 (en) Graph convolutional neural network and knowledge base-based paper correlation analysis method
Cao et al. Rankcompete: Simultaneous ranking and clustering of information networks
Mahalakshmi et al. Ensembling of text and images using deep convolutional neural networks for intelligent information retrieval
Meng et al. Concept-concept association information integration and multi-model collaboration for multimedia semantic concept detection
Zhu et al. Multimodal sparse linear integration for content-based item recommendation
Bu et al. Unsupervised face-name association via commute distance
Escalante et al. An energy-based model for region-labeling

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
EE01 Entry into force of recordation of patent licensing contract
EE01 Entry into force of recordation of patent licensing contract

Application publication date: 20200110

Assignee: ZHEJIANG ANDA SYSTEM ENGINEERING Co.,Ltd.

Assignor: HANGZHOU DIANZI University

Contract record no.: X2022980022900

Denomination of invention: Hybrid recommendation method based on graph convolution neural network

Granted publication date: 20211015

License type: Common License

Record date: 20221124

EE01 Entry into force of recordation of patent licensing contract
EE01 Entry into force of recordation of patent licensing contract

Application publication date: 20200110

Assignee: Zhejiang haomeng Technology Co.,Ltd.

Assignor: HANGZHOU DIANZI University

Contract record no.: X2022980024264

Denomination of invention: Hybrid recommendation method based on graph convolution neural network

Granted publication date: 20211015

License type: Common License

Record date: 20221202