K-means algorithm with improved initialization for clustering fruit plants

Abstract

Abstract Clustering is one of data mining technique that can make a set of objects in such way that objects in the same group are more similar in some particular manner to each other than to those in other groups. K-means algorithm is one of clustering methods. Standard K- means algorithm has fast and simple computation, but it has the pitfall of randomly choosing initial the center of cluster. In this paper, we propose a mean method combined with interval index based on number cluster to choose initial the center of cluster. It can eliminate the randomness of the selection of initial the center of cluster, so it can find the optimum the center of cluster faster. The effectiveness of algorithm can be seen by maximum iteration of each algorithm. And fruit plants data will be used as data test.