A network community graph generation method, device, computer equipment and storage medium

By introducing the first and second reference thresholds in the network community graph generation, determining the community attributes of the nodes and building a network subgraph, the problems of existing algorithms being limited in resolution and unstable results in large-scale networks are solved, and more efficient and accurate network community graph generation is achieved.

CN114741383BActive Publication Date: 2025-05-09TENCENT TECHNOLOGY (SHENZHEN) CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Application Number
CN202110021879.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-01-07
Publication Date
2025-05-09
Estimated Expiration
2041-01-07

AI Technical Summary

Technical Problem

Existing community discovery algorithms, such as the Louvain algorithm and the LPA algorithm, have the problem of resolution limitation when dealing with large-scale networks, and cannot identify communities with small nodes, and randomness may lead to unstable results.

Method used

By obtaining the correlation parameters between the pending network map and its nodes, using the first reference threshold and the second reference threshold, the community attributes of the nodes are determined and the network sub-graph is constructed, thereby generating a more accurate and stable network community map.

Benefits of technology

The efficiency and accuracy of network community graph generation is improved, the problem of resolution limitation is reduced, and the instability of the result is avoided by introducing a threshold mechanism.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114741383B_ABST
    Figure CN114741383B_ABST
Patent Text Reader

Abstract

The embodiment of the present application proposes a network community generation method, device, computer equipment and storage medium, the method is: obtaining a network graph to be processed containing multiple node objects, and obtaining an association parameter between any two node objects in the multiple node objects; determining the community attribute of each first node object in the multiple node objects according to the association parameter between any two node objects and a first reference threshold; constructing at least one network subgraph corresponding to the network graph to be processed according to the association parameter between any two node objects and a second reference threshold; constructing a network community graph corresponding to the network graph to be processed according to the community attribute of each first node object, the association parameter between any two node objects and at least one network subgraph. Through the present application, the efficiency and accuracy of network community graph generation can be improved.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Data processing method and device, storage medium and computer device

    CN110765320A

  • Personnel relationship analysis method and device and readable storage medium

    CN111177473A