An Optimum Method for Enhancing the Computational Complexity of K-Means Clustering Algorithm with Improved Initial Centers

A. Mallikarjuna Reddy, Ramapuram Gautham

Abstract: The amount of data stored in databases continues to grow fast. Intuitively; this large amount of stored data contains valuable hidden patterns; which could be used to improve the decision-making process. Data mining is a process of identifying specific patterns from large amount of data. In data mining; Clustering is one of the major data analysis methods and the K-means clustering algorithm is widely used for many practical applications. Though it is widely used; its generates a local optimal solution based on the randomly chosen initial centroids and the computational complexity is very high O (nkl). In order to improve the performance of the K-means algorithm several methods have been proposed in the literature. The proposed algorithm enhances the performance of K-means clustering algorithm. This algorithm consists of two phases. Phase I algorithm finds the better initial centroids; Phase II algorithm is used for the effective way of assigning data points to suitable clusters. Experiments on a number of real-world data sets show that the proposed approach has produces consistent clusters compared to some well-known methods; reducing the computational complexity O (nlogn) of k-means algorithm. Though the proposed method will improve the accuracy and efficiency of k-means clustering algorithm.

Keywords: Mining, Clustering, Knowledge Discovery in Databases, K-means clustering algorithm, Optimum method