PSO Clustering Algorithm Based on Cooperative Evolution |
| |
Authors: | QU Jian-hua SHAO Zeng-zhen LIU Xi-yu |
| |
Institution: | [1]School of Management and Economics, Shandong Normal University, Jinan 250014, China [2]School of Information Science and Engineering, Shandong Normal University, J inan 250014, China |
| |
Abstract: | Among the bio-inspired techniques, PSO-based clustering algorithms have received special attention. An improved method named Particle Swarm Optimization (PSO) clustering algorithm based on cooperative evolution with multi-populations was presented.It adopts cooperative evolutionary strategy with multi-populations to change the mode of traditional searching optimum solutions. It searches the local optimum and updates the whole best position (gBest) and local best position (pBest) ceaselessly. The gBest will be passed in all sub-populations. When the gBest meets the precision, the evolution will terminate. The whole clustering process is divided into two stages. The first stage uses the cooperative evolutionary PSO algorithm to search the initial clustering centre.The second stage uses the K-means algorithm. The experiment results demontrate that this method can extract the correct number of clusters with good clustering quality compared with the results obtained from other clustering algorithms. |
| |
Keywords: | Particle Swarm Optimization ( PSO ) clustering algorithm cooperative evolution multi-populations |
本文献已被 CNKI 维普 万方数据 等数据库收录! |