Method, apparatus and electronic device for generating node representation in heterogeneous graph
A heterogeneous graph and node technology, applied in the field of Internet and machine learning, can solve the problems of heterogeneous graph structure information loss, poor node representation accuracy, etc., to achieve the effect of ensuring no loss, improving accuracy, and ensuring integrity
Active Publication Date: 2022-04-22
BEIJING BAIDU NETCOM SCI & TECH CO LTD
View PDF6 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] This application proposes a method, device and electronic device for generating node representations in heterogeneous graphs, which are used to solve the problem of using meta-path sampling methods to train heterogeneous graphs as isomorphic graphs in the prior art, resulting in the structure of heterogeneous graphs Information loss, technical issues with less accurate representation of generated nodes
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0061] Exemplary embodiments of the present application are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present application to facilitate understanding, and they should be regarded as exemplary only. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the application. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.
[0062] The method, device and electronic device for generating node representations in heterogeneous graphs of the present application are described below with reference to the accompanying drawings.
[0063] A large number of problems in the real world can be abstracted into a graph model, that is, a collection of nodes and edges, from knowledge graphs to probabilis...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The application discloses a method, device and electronic equipment for generating node representations in heterogeneous graphs, and relates to the technical field of machine learning. The specific implementation scheme is: obtain a heterogeneous graph; input the heterogeneous graph into a heterogeneous graph learning model to generate node representations of each node in the heterogeneous graph, and the heterogeneous graph learning model generates the described heterogeneous graph through the following steps Node representation of each node: split the heterogeneous graph into multiple subgraphs, each subgraph includes two types of nodes and an edge type between the two types of nodes; generate each node according to multiple subgraphs node representation. The scheme of the present application can obtain the structural information of the following graphs with different edge types, ensure that the structural information of the heterogeneous graph is not lost, improve the accuracy of node representation, and solve the problem of treating heterogeneous graphs as isomorphic graphs by using meta-path sampling methods in the prior art Training results in the loss of structural information of heterogeneous graphs, and the technical problems of poor accuracy of generated node representations.
Description
technical field [0001] The present application relates to the technical fields of the Internet and machine learning, and in particular to a method, device and electronic device for generating node representations in heterogeneous graphs. Background technique [0002] A large number of problems in the real world can be abstracted into a graph model, that is, a collection of nodes and edges. For example, the relationship between each user and other users in a social platform can be abstracted into a graph model. Each node in the graph model can be expressed in a vector form, which can be applied to a variety of downstream tasks, such as node classification, link prediction, community discovery, etc. [0003] At present, in heterogeneous graph node representation learning, different walk sequences are obtained through meta-path sampling, and the walk sequences are treated as sentence sequences through word2vec and other training methods to train the walk sequences to obtain gra...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06N20/00G06F16/901
CPCG06N20/00G06F16/9024G06N5/022
Inventor 李伟彬朱志凡苏炜跃何径舟冯仕堃曹宇慧陈徐屹朱丹翔
Owner BEIJING BAIDU NETCOM SCI & TECH CO LTD