Author(s): Shivaprasad B, Dr. R.V.Krishnaiah
Published in: International Journal of Engineering Research & Technology
License: This work is licensed under a Creative Commons Attribution 4.0 International License.
Volume/Issue: Vol.1 - Issue 9 (November - 2012)
In data mining domain, high-dimensional and correlated data sets are used frequently. Working with high-dimensional data for data mining operations like clustering has become a common task in data mining. In this paper, we consider similarity search approaches in the presence of high-dimensional data. The existing indexing approaches such as vector approximation has some drawbacks such as ignoring dependencies across dimensions. This results in sub optimality in results. However, clustering makes use of inter- dimensional correlations and can represent a dataset used. Pruning clusters that are not relevant is done by existing algorithms. However, they are based on bounding rectangles, bounding hyper spheres and they lack in efficiency in nearest neighbor search. We propose a new algorithm for separating hyperlange boundaries of Voronoi clusters. This is known as cluster-adaptive distance bound algorithm which is complements cluster based index. It performs spatial filtering well besides reducing the storage overhead. Our method can be used with Mahalanobis and Euclidean similarity measures. We developed a prototype application for demonstrating the efficiency of the proposed method. The results revealed that the proposed method is effective and can be used in the real world data mining applications.
Number of Citations for this article: Data not Available
7 Paper(s) Found related to your topic: