Social networks can be viewed as a reflection of the real world which can be studied to gain insight into the real life societies and events. During the last decade, community detection as a fundamental part of social network analysis has been explored widely, however because of the complex nature of the network, it is still an open problem. In this paper, we propose a knowledge-based evolutionary algorithm to solve this problem by using a multi-population cultural algorithm. In our algorithm, knowledge is extracted from the network to guide the search direction and find the optimal solution. Meanwhile, in each step, the knowledge is updated based on the current states of the network. The results of comparison between our method and other well-known algorithms show that our algorithm is capable to find the true communities faster and more accurately than the others.
