Research And Application On Ant Colony Clustering Algorithm

Data mining is the process of automatically searching large volumes of data for patterns and rules. Clustering is an important technique of data mining,and also an efficient method for people understanding and s eeking the internal relations of things.It can be used as the tool of data mining to discover the in-deep information of data distribution in database .Otherwise, it is considered as the pretreatment process of other data mining algorithms,it is widely used in fields of engineering and technology. Clustering divides data objects into different clusters so that the elements atached to the diferent clusters have dissimilarities and the ones attached to the same clusters have similarities.On the research of the algorithm based on ACA and clustering algorithm combination,I have sufficiently studied the basic principle and capability of the existing ant colony algorithms. I presents a algorithm that combines the object and the environment arount it together and then decides to pick it or not.It not only can avoid man-made datas about the number the the clustering,but also can aoid local optimum.At last I apply it in the partition of client in the security trade.It provides a new thought and approach for the clustering research. Thus, researches of this paper have the great significance in theory and practice.