A multiobjective discrete bat algorithm for community detection in dynamic networks

Some evolutionary based clustering approaches for community detection in dynamic networks need an input parameter to control the preference degree of snapshot and temporal cost. To break the limitation of parameter selection and improve the quality of detecting communities in dynamic network further, a multiobjective discrete bat algorithm (MDBA) is proposed to detect community structure in dynamic networks in this paper. In the proposed algorithm, the bat location updating strategy is designed in discrete form. In addition, turbulence operation and mutation strategy are presented to guarantee the diversity of the population. The non-dominated sorting and crowding distance mechanism are used to keep good solutions during the generation. The experimental results both on synthetic and real networks show that MDBA algorithm is competitive and will get higher accuracy and lower error rate than the compared algorithms.