Fruit Fly K-Means Clustering Algorithm

Abstract

Clustering is one of the main data mining tasks. It aims to grouping the data objects into significant clusters such that the similarity of objects within clusters is maximized, and the similarity of objects from different clusters is minimized. K-Means algorithm is most commonly used algorithm for unsupervised clustering problem. But it has some problems which make it unreliable. Initialization of the random cluster centers, number of clusters and terminating condition play a major role in quality of clustering achieved. In this paper we proposed Fruit Fly algorithm to select the initial centroids for K-Means algorithm in order to optimize the number of clusters. The experimental analysis is conducted on Cocaine dataset to validate the proposed method.