Open AccessA Multi-resolution Community Partition Method based on Node Degree
Author(s) -
Xia Qin,
Fan Wu,
Kebin Chen
Publication year - 2021
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1924/1/012011
Subject(s) - degree (music) , partition (number theory) , node (physics) , computer science , similarity (geometry) , resolution (logic) , complex network , algorithm , data mining , community structure , network partition , mathematics , artificial intelligence , statistics , image (mathematics) , distributed computing , world wide web , engineering , physics , structural engineering , combinatorics , acoustics
Community detection is one of the significant methods to study the properties of complex networks. Community resolution, as an important index of detection algorithm, is often used to evaluate the performance of the algorithm, but many classical algorithms have shortcomings in this respect. Based on this, this paper proposes a multi-resolution community partition method based on node degree. By introducing similarity coefficient and propagation coefficient of degree, the resolution of network is effectively controlled. Experiments on a scale-free network with 100 nodes show that the proposed algorithm is stable and efficient, with low time complexity. The number of communities can vary from 3 to 27 with different combinations of two coefficients, which can well discover the rules and characteristics of the network.