social network clustering with genetic algorithm

Many people today establish part of their relationship with friends through virtual social networks. One of the most practical issues in computer science is the issue of data clustering, which has many applications in the field of social networking, pattern finding, and data similarity. Many researchers in various fields have done various researches about it. On the other hand, the possibility of modeling many problems has caused widespread attention to graph clustering. Since single-objective optimization algorithms can not optimize all the objectives of community discovery, in this research, a two-objective meta-heuristic algorithm is proposed for this purpose. Researchers have used several genetic algorithms to identify communities, but the proposed algorithm uses two goals together, which form the basis of defining communities, which improves efficiency and accuracy. The performance results of the proposed method are compared with other genetic-based algorithms by standard data sets in the field of social network analysis and the results show the superiority of the proposed method over other methods.

Keywords:

Social Networks , Optimization Algorithms , Two-Objective Genetic Algorithm , Modularity , Pareto Front , Fitness Function