A Modified K-means Algorithms - Bi-Level K-Means Algorithm

Abstract

In this paper, a modified K-means algorithm is proposed to categorize a set of data into smaller clusters. Kmeans algorithm is a simple and easy clustering method which can efficiently separate a huge number of continuous numerical data with high-dimensions. Moreover, the data in each cluster are similar to one another. However, it is vulnerable to outliers and noisy data, and it spends much executive time in partitioning data too. Noisy data, outliers, and the data with quite different values in one cluster may reduce the accuracy rate of data clustering since the cluster center cannot precisely describe the data in the cluster. In this paper, a bi-level K-means algorithm is hence provided to solve the problems mentioned above. The bi-level K-means algorithm can give an expressive experimental results.