Authors: Yifei Yue; Chaokun Wang; Xiang Ying; Jun Qian
Addresses: School of Software, Tsinghua University, Beijing 100084, China ' School of Software, Tsinghua University, Beijing 100084, China ' School of Software, Tsinghua University, Beijing 100084, China ' School of Software, Tsinghua University, Beijing 100084, China
Abstract: Network data plays an important role in biological research. For example, the interaction between proteins in living cells forms large complex networks. The corporation of cells in a living body also makes up networks. As an important approach to analysing the topology of network data, community detection methods have attracted a great interest of researchers, and different algorithms have been developed during the past decade. However, the diversity of these algorithms also makes users confused to choose a suitable one according to the specific application. In this paper, we present CoDeT, a system which integrates 11 state-of-the-art community detection algorithms and 12 recognised metrics, to address the difficulty. Especially, CoDeT is capable to recommend the most suitable algorithm for users when they consider multiple algorithms for a given data set. Experimental results show that the recommended algorithms by our system are effective on bioinformatic networks. In addition, with our provided C++, Python and web service interfaces, users can easily select the most convenient one to start their experience.
Keywords: community detection; bioinformatic network analysis; algorithm recommendation.
International Journal of Data Mining and Bioinformatics, 2017 Vol.19 No.1, pp.52 - 74
Received: 26 Jul 2017
Accepted: 26 Jul 2017
Published online: 02 Dec 2017 *