期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2015
卷号：3
期号：4
DOI：10.15680/ijircce.2015.0304068
出版社：S&S Publications
摘要：In this paper to focuses on discovery of functional group of the connectivity atom for drug effects ofchemical compound structured data with position of each atom. A simple Kmeans algorithm, select an initial centroiddistance randomly for analysing the data. In the proposed method an Enhanced K means algorithm, forms a functionalgroup of inter connected atoms based on calculate initial centroid distance instead of random selected. Thepharmaceutical compounds specifically represented as atom number, atom name like carbon, hydrogen, nitrogen,oxygen with connected atoms. Here it can be experimented the number of iterations are reduced and performance oftime accuracy can improve when compare with chameleon and Birch algorithm.
