Conference proceeding
Community Detection in Graphs through Correlation
PROCEEDINGS OF THE 20TH ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING (KDD'14), pp.1376-1385
01/01/2014
DOI: 10.1145/2623330.2623629
Abstract
Community detection is an important task for social networks, which helps us understand the functional modules on the whole network. Among different community detection methods based on graph structures, modularity-based methods are very popular recently, but suffer a well-known resolution limit problem. This paper connects modularitybased methods with correlation analysis by subtly reformatting their math formulas and investigates how to fully make use of correlation analysis to change the objective function of modularity-based methods, which provides a more natural and effective way to solve the resolution limit problem. In addition, a novel theoretical analysis on the upper bound of different objective functions helps us understand their bias to different community sizes, and experiments are conducted on both real life and simulated data to validate our findings.
Details
- Title: Subtitle
- Community Detection in Graphs through Correlation
- Creators
- Lian Duan - New Jersey Institute of TechnologyWillian Nick Street - University of IowaYanchi Liu - New Jersey Institute of TechnologyHaibing Lu - Santa Clara University
- Resource Type
- Conference proceeding
- Publication Details
- PROCEEDINGS OF THE 20TH ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY AND DATA MINING (KDD'14), pp.1376-1385
- Publisher
- Assoc Computing Machinery
- DOI
- 10.1145/2623330.2623629
- Number of pages
- 10
- Language
- English
- Date published
- 01/01/2014
- Academic Unit
- Nursing; Business Analytics; Computer Science; Bus Admin College
- Record Identifier
- 9984380500102771
Metrics
10 Record Views