A Multimodal Optimization and Surprise Based Consensus Community Detection Algorithm

Guanbo Jia, Shan He, Zexuan Zhu, Jing Liu, Ke Tang

Abstract

A new community structure measure called Surprise has been proposed to address the resolution limit problem of modularity. However, our analysis shows that, similar to modularity, Surprise also suffers from the so-called extreme degeneracy problem, which leads to unstable module identification results. To solve this problem, we propose a novel Multimodal Optimization and Surprise based Consensus Community Detection (MOSCCoD) algorithm. Experimental results show that MOSCCoD has overcome the extreme degeneracy problem of Surprise and shown a very competitive performance in terms of stability and accuracy.
BibTeX
@inproceedings{jia2015multimodal,
  author = {Guanbo Jia and Shan He and Zexuan Zhu and Jing Liu and Ke Tang},
  title = {{A Multimodal Optimization and Surprise Based Consensus Community Detection Algorithm}},
  booktitle = {GECCO (Companion)},
  year = {2015},
  doi = {10.1145/2739482.2764656},
}

Loading related content...

Figure