A New Cell-Based Clustering Method for High-Dimensional Data Mining Applications

Abstract

Many clustering methods are not suitable for high-dimensional data mining applications because of the so-called ‘curse of dimensionality’ and the limitation of available memory. In this paper, we propose a new cell-based clustering method for the high-dimensional data mining applications. The proposed clustering method provides efficient cell creation and cell insertion algorithms using a space-partitioning technique, as well as makes use of a filtering-based index structure using an approximation technique. In addition, we compare the performance of our cell-based clustering method with the CLIQUE method which is well known as an efficient grid-based clustering method for high-dimensional data. The experimental results show that our clustering method achieves better performance on cluster construction time and retrieval time.