Improved the Efficiency of Self Optimal Clustering Technique using Particle Swarm Optimization

Abstract: In the last decade, various methods able to detect multiple clustering solutions have been introduced. According to the survey, they can briery be categorized into methods operating on the original data-space, methods performing space transformations, and methods analysing subspace projections. The main idea is to consider each subspace as a multiple fitness constraint. For the performance evaluation of proposed algorithm used three real time dataset from UCI machine learning center. The proposed algorithm implemented in Matlab software and measures some standard parameter for the validation of proposed methodology. Our proposed method compares with two well know clustering technique such as K-means, FCM and SOC algorithm. Results shows better performance of proposed algorithm compared in existing these two algorithms.

Keywords: Clustering, GA, SOC, PSO.