Ant Colony Algorithm Research And Its Application On Clustering Of Web Mining

Today's human society is an information society. Because of science and technology especially network technology, World Wide Web is developing in breadth and depth. People have been drowning in information. These massive and heterogeneous information resources contain large numbers potential knowledge. How to find out the credible, effective and useful information from these massive data sources becomes a major research focus in our society today. Data mining technology has been developed vigorously in this background. This paper is the study of ant colony clustering algorithm in Web mining analysis of clustering, improving the existing ant colony clustering algorithm to get better performance and to be more suitable for Web Mining Cluster Analysis . The following will be included in this paper:First of all, a general overview of data mining is given. Some of the basic concepts of data mining, data mining systems and some basic knowledge of data mining are introduced,then the cluster analysis of data mining techniques are described as well as the basic concepts of clustering, and the current status of clustering research, as well as several important clustering methods will also be analyzed and compared.Then ant colony algorithm is summarized. The basic model of ant clustering algorithm and LF ant colony clustering algorithm which builds on the basic model will be introduced. The paper will analyze the strengths and weaknesses of the LF ant colony clustering algorithms, then proposes an improved ant colony clustering algorithm based on LF algorithm. Through simulation experiments, it is proved that the performance of the improved algorithm is better than the LF algorithm. Our improved algorithm has solved some problems of the LF algorithm such as slow convergence, long iterative cycle and so on. At last a small cluster system is designed and implemented to verify our improved algorithm.