| The purpose of Data Mining is to abstract potential, valuable knowledge and useful information from plentiful data. Cluster analysis is one of the research domains of data mining,which has important appliances in many domains such as in business, biology,medicine, geography, web archive, and it becomes one of the hot research problems.As a new form of Artificial Intelligence, Swarm Intelligence is becoming a hot domain in Artificial Intelligence and relevant fields. At present,more researches on swarm intelligence algorithm mainly include particle swarm optimization(PSO) algrithm and ant colony algorithm. In this paper,we have studied the problems existing in the two kinds of swarm intelligent algorithm and applied the algorithm to cluster analysis successfully.(1) The particle swarm optimization algorithm is improved by parameters and applied to the cluster analysis.The cluster algorithm based on parallel particle swarm optimizer using adaptive inertia weight is proposed in this paper,which enhances the capabilities of finding global optimal solution,convergence rate, efficiency of convergence and so on. Theoretical analysis and experiments show that the proposed algorithm is obviously superior to k-means cluster algorithm,genetic cluster algorithm and basic pso algorithm.(2) The ant colony optimization algorithm is improved on the basis of based on the basic model of ant colony optimization proposed by Deneubourg,LF ant colony algorithm by Lumer and Faieta,and swarm intelligence cluster algorithm using of information entropy theory by liu bo.The proposed algorithm in this paper not only improves the speed and quality of clustering,but also solves the problems of more parameters of LF algorithm and not being applied to continuous attributes. Contrast to the traditional k-means algorithm,the algorithm has the features of dealing with the isolated dot effectively, having better anti-noise capability,and not needing to set the number of cluster. |
PDF Full Text Request