Dynamic sub-graph generation method and dispute detection method based on node features and reply paths

A detection method and node technology, which is applied in complex mathematical operations, biological neural network models, data processing applications, etc., can solve the problems of model performance impact, inability to model dispute modes, inability to pay attention to the overall information of local subgraphs, etc., to improve The effect of detection performance

Pending Publication Date: 2021-08-13
中科计算技术创新研究院 +1
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Benefits of technology

This patented technology allows researchers to create an algorithm called Minimax which uses dynamic generative methods to detect disputed areas within a dataset. It suggests generating multiple versions of this area by combining it from both original nodes or relying upon existing ones. These modifications are then used to infer new sections about those regions where conflicting topics have been identified during investigation. By doing these steps over time, scientists could identify issues more accurately than previously possible without having complete datasets containing all relevant details.

Problems solved by technology

Technological Problem addressed in this patented technical problem involves detecting potential conflicts within a community where users may want to join together without being influenced negatively towards one another. Current methods require manual input from experts who manually analyze large amounts of evidence collected during investigations into relevant areas.

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Dynamic sub-graph generation method and dispute detection method based on node features and reply paths
  • Dynamic sub-graph generation method and dispute detection method based on node features and reply paths
  • Dynamic sub-graph generation method and dispute detection method based on node features and reply paths

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0064] This embodiment is a method for mining key local disputes for dispute detection based on a dynamic subgraph generation method, including a dynamic subgraph generation method and a local dispute mining method based on node characteristics and reply paths.

[0065] A subgraph centered on a node (eg, node i) should include local discussions related to it, which exist in the reply path of the central node i. The dynamic subgraph generation method in this embodiment first calculates the correlation between each node in the reply path and the central node i, and then truncates the path to remove irrelevant nodes. The collection of all truncated paths constitutes the subgraph corresponding to node i, which is the relevant local discussion.

[0066] Ordinary GCN models use adjacency matrix, degree matrix, and Laplacian matrix to model the interaction between nodes and first-order neighbor nodes. In this example, path matrix, path length matrix, and path Laplacian matrix are used t

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

No PUM Login to view more

Abstract

The invention relates to a dynamic subgraph generation method based on node features and reply paths, and a dispute detection method. The dynamic subgraph generation methodincludes the following steps: S1, constructing a path matrix P and a path length matrix S based on a post-comment graph G, wherein the path matrix P records all paths from each node to a terminal node in the graph G, the terminal nodes comprise post nodes in the graph G and comment nodes without replies, and the path length matrix S records the length of each path in the path matrix; s2, calculating a path Laplacian matrix L based on the path matrix P and the path length matrix S; s3, on the basis of the path Laplacian matrix L and the content features of the nodes in the graph G, calculating and obtaining the expression of the current node sensing path information; and S4, on the basis of the similarity between the current node and all nodes on the corresponding path, and reserving the most important part of nodes on each path, wherein the important nodes on all the paths form a sub-graph corresponding to the current node, and the nodes in the sub-graph are local discussion related to the current node.

Description

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

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
no application Login to view more
Owner 中科计算技术创新研究院
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products