Density based clustering technique for efficient data mining
Ashikur Rahman, A.K.M. Rasheduzzaman Chowdhury, Daud Jamilur Rahman, Abu Raihan Mostofa Kamal · 2008
Clustering analysis is an important function of data mining. There are various clustering methods in Data Mining. Based on these methods various clustering algorithms are developed. A recent approach for clustering analysis is based on “Swarm Intelligence”. Based on this “Swarm Intelligence” an algorithm was proposed named “Ant-Cluster algorithm”. However, existing “Ant clustering” algorithm has a limitation in finding the value of two constant K1and K2, which is user defined., for computing the value of the picking up probability Pp and dropping probability Pd. In this paper our approach is to gain the value of Pp and Pdwithout giving the user defined value of K1and K2. We also intend to retain the Pp and Pdin between 0 to 1 in order to get optimized result.