To search, Click below search items.


All Published Papers Search Service


An automatic Clustering Method based on Maximum Distance


Hongbo Zhou, Yongqiang Feng, Juntao Gao


Vol. 15  No. 4  pp. 40-43


Traditional k-means clustering algorithm needs to determine cluster number and select initial cluster centers in advance,considering the shortcomings of k-means algorithm,the paper proposed an improved efficient clustering algorithm.The algorithm does not require pre-determined number of clusters and initial cluster centers,select cluster centers according to the principle of maximum distance,divide cluster according to the principle of minimum distance,and determine the optimal cluster partition according to the distance evaluation function.The improved algorithm avoids the choice of the value of cluster number and initial centers.Hence this method can produce more accurate clustering results than the standard k-means algorithm.Experimental results show that the improved algorithm has good performance and high time efficiency.


Euclidean distance, cluster center, maximum distance principle, minimum distance principle, distance evaluation function