An Improved Community Detection Algorithm Based on DCT and K-Means

DOI

10.2991/cimns-16.2016.73How to use a DOI?

Keywords

community structure; DCT; k-means; curse of dimensionality

Abstract

Detecting an overlapping and hierarchical community structure can give a significant insight into structural and functional properties in complex networks. In this paper, we propose an improved algorithm to detect communities in the complex network. The proposed algorithm use discrete cosine transform (DCT) to transfer the topology information into frequency domain, and reduce the dimension of frequency signal with a preliminary threshold, and at last cluster the nodes using K-means. Finally, we apply the proposed algorithm in the real and artificial networks. The simulation results show that the proposed algorithm can avoid curse of dimensionality and it is more accurate than some existing mechanisms.

Copyright

© 2016, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Lin Li
AU  - Kefeng Fan
AU  - Jiezhong Gong
AU  - Hao Peng
PY  - 2016/09
DA  - 2016/09
TI  - An Improved Community Detection Algorithm Based on DCT and K-Means
BT  - Proceedings of the 2016 International Conference on Communications, Information Management and Network Security
PB  - Atlantis Press
SP  - 293
EP  - 297
SN  - 2352-538X
UR  - https://doi.org/10.2991/cimns-16.2016.73
DO  - 10.2991/cimns-16.2016.73
ID  - Li2016/09
ER  -