Based On Cluster Analysis Of The Data Mining Algorithm

Data Mining is a hotspot, which various domains are studying and attracts lots of pursuers. Clustering analyzing is a function of Data Ming, in other words, it is a study branch of DM and now its most study is focused on clustering algorithm, the main purpose is to produce practical algorithms with better performance. In this paper, at first I present the aim of Data Mining and status quo of investigation as well as its main study contents simply. The primary purpose of above introduction of Data Mining is to make people acquaint himself with DM. In succession based on the brief explanation of Clustering Analyzing, I completed analyzing of representative and leading algorithms of Clustering in existence and digged out their advantages and disadvantages as well as theirs condition in point. Moreover, aimed at the disadvantages of partition method's representative algorithms, I put forward a new algorithm named Distribution Based Clustering of LArge Database(DBCLAD). It avoids the awkard conditions that need user to provide parameters that is difficult to decide. What's more it can discover arbitrary shape cluster. So it is promising, all are the core of the paper. At last I give the realization of this algorithm and apply it to an instance.