ABSTRACT
One important aspect of graphs representing complex systems is community (or group) structure---assigning vertices to groups, which have dense intra-group connections and relatively sparse inter-group connections. Community detection is of great importance in various domains, where real-world complex systems are represented as graphs, since communities facilitate our understanding of the graph and thus of the underlying system. However, this is known to be a hard optimization problem.
In this study we pursue the following question: Have Evolutionary Computation-Based Methods proven their worth for this complex domain, or is it currently better to rely on other state-of-the-art methods? While several works compare state-of-the-art methods for community detection (see [8] and [11] for recent surveys), we are unaware of other attempts to compare methods based on evolutionary computation to other methods.
After describing some recent algorithms for this problem, and comparing them in various ways, we conclude that evolutionary computation-based method for community detection have indeed developed to hold their own against other methods for several variants of this problem. However, they still need to be applied to more difficult problems and improve further to make them in par with other methods.
- J. Chen and Y. Saad. Dense subgraph extraction with application to community detection. IEEE Trans. Knowl. Data Eng., 24(7):1216--1230, 2012. Google ScholarDigital Library
- P. Gopalan and D. M. Blei. Modeling overlapping communities with node popularities. In Advances in Neural Information Processing Systems 26, 2013, Nevada, United States., pages 2850--2858, 2013.Google Scholar
- P. Jiang and M. Singh. Spici: a fast clustering algorithm for large biological networks. Bioinformatics, 26(8):1105--1111, 2010. Google ScholarDigital Library
- C. Liu, J. Liu, and Z. Jiang. A multiobjective evolutionary algorithm based on similarity for community detection from signed social networks. IEEE Trans. Cybernetics, 44(12):2274--2287, 2014.Google ScholarCross Ref
- J. Liu, W. Zhong, H. A. Abbass, and D. G. Green. Separated and overlapping community detection in complex networks usingmultiobjective evolutionary algorithms. In Proceedings of the IEEE Congress on Evolutionary Computation, CEC 2010, Barcelona, Spain, pages 1--7, 2010.Google Scholar
- K. Macropol and A. K. Singh. Scalable discovery of best clusters on large graphs. PVLDB, 3(1):693--702, 2010. Google ScholarDigital Library
- C. Pizzuti. A multi-objective genetic algorithm for community detection in networks. In ICTAI 2009, 21st IEEE International Conference on Tools with A.I., New Jersey, USA, pages 379--386, 2009. Google ScholarDigital Library
- B. Saha, A. Mandal, S. B. Tripathy, and D. Mukherjee. Complex networks, communities and clustering: A survey. CoRR, abs/1503.06277, 2015.Google Scholar
- C. Shi, Y. Cai, D. Fu, Y. Dong, and B. Wu. A link clustering based overlapping community detection algorithm. Data Knowl. Eng., 87:394--404, 2013. Google ScholarDigital Library
- J. Su and T. C. Havens. Fuzzy community detection in social networks using a genetic algortihm. In Fuzzy Systems (FUZZ-IEEE), 2014 IEEE International Conference on, pages 2039--2046, July 2014.Google ScholarCross Ref
- P. Wadhwa and M. Bhatia. Community detection approaches in real world networks: A survey and classification. IJVCSN, 6(1):35{51, 2014.Google Scholar
- J. Xie and B. K. Szymanski. Towards linear time overlapping community detection in social networks. In Advances in Knowledge Discovery and Data Mining 2012. Google ScholarDigital Library
Index Terms
- Are Evolutionary Computation-Based Methods Comparable to State-of-the-art non-Evolutionary Methods for Community Detection?
Recommendations
Community-based anomaly detection in evolutionary networks
Networks of dynamic systems, including social networks, the World Wide Web, climate networks, and biological networks, can be highly clustered. Detecting clusters, or communities, in such dynamic networks is an emerging area of research; however, less ...
Hybrid biogeography-based evolutionary algorithms
Hybrid evolutionary algorithms (EAs) are effective optimization methods that combine multiple EAs. We propose several hybrid EAs by combining some recently-developed EAs with a biogeography-based hybridization strategy. We test our hybrid EAs on the ...
Analysis of evolutionary multi-tasking as an island model
GECCO '18: Proceedings of the Genetic and Evolutionary Computation Conference CompanionRecently, an idea of evolutionary multi-tasking has been proposed and applied to various types of optimization problems. The basic idea of evolutionary multi-tasking is to simultaneously solve multiple optimization problems (i.e., tasks) in a ...
Comments