K-MEANS CLUSTERING WITH SWARM OPTIMIZATION FOR SOCIAL NETWORK COMMUNITY DETECTION

Purchase PDF

Published: 2015-04-14

Page: 220-230


YOMNA M. EL BARAWY *

Al-Azhar University, Faculty of Science, Cairo, Egypt.

LAMIAA M. EL BAKRAWY

Al-Azhar University, Faculty of Science, Cairo, Egypt.

NEVEEN I. GHALI

Al-Azhar University, Faculty of Science, Cairo, Egypt.

*Author to whom correspondence should be addressed.


Abstract

Now any organization success is depending on its network environment i.e the ability to understand the relation between the actors of this organization which is the work of social network analysis (SNA).SNA has many applications in business eld as customer behavior, community development, communication studies and marketing. Community detection is one of the SNA elds which shows actors who interact with each other more than ones outside their group. This paper present the idea of using the output of an optimization algorithms Particle Swarm Optimization (PSO) and Exponential Particle Swarm Optimization EPSO as input to the k-means clustering algorithm in order to have a well community detection for social network data. It deals with the community detection problem as a clustering one. The Experimental results show that using EPSO algorithm to optimize cluster centroids is more ecient than using PSO. Since it gives a better tness value and take less time as the size of the dataset increase.

Keywords: Social network community detection, social network analysis, particle Swarm Optimi-zation, exponential particle swarm optimization, K-means clustering


How to Cite

EL BARAWY, Y. M., EL BAKRAWY, L. M., & GHALI, N. I. (2015). K-MEANS CLUSTERING WITH SWARM OPTIMIZATION FOR SOCIAL NETWORK COMMUNITY DETECTION. Asian Journal of Mathematics and Computer Research, 3(4), 220–230. Retrieved from https://ikprress.org/index.php/AJOMCOR/article/view/40

Downloads

Download data is not yet available.