Abstract

Genetic algorithms have been used in community detection due to their efficiency and accuracy in automatic discovery of communities in complex networks. The traditional method of population initialization for genetic algorithms does not fully consider topology of a network. Therefore, the quality of the initial population may be poor, which may consequently slow down the convergence of the entire process. According to the characteristics of social networks, we propose a K-path initialization method which makes full use of the topological information of a given network. The main focus of our study is to find whether a K-path initialized generic algorithm can bring significant increase in Q value after the FIRST iteration, and to prove whether such algorithm can accelerate the convergence of the entire process for faster community detection over a randomly initialized genetic algorithm. By applying this new algorithm to Karate, Football, and Jazz, we found that the K-path initialized algorithm can increase the Q value by 50---160% on average over that without using K-path initialization after the first round of iteration. This initial advantage is then accumulated during the subsequent iterations by reducing the number of total iterations by at least 28% and at most 41%.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.