Performance Analysis & Comparison b/w Enhanced K-Means & Orthogonal Partitioning (OC), based on proposed New Approach: "DRID"

Summary Clustering provides grouping together of similar data items. This technique provides a high level view of the database. Clustering technique is a technique that merges and combines techniques from different disciplines such as mathematics, physics, mathprogramming, statistics, computer sciences, artificial intelligence and databases etc. Variety of clustering algorithms exists and belongs to several different categories. Two Prominent Categories are Distance based and Grid based (example: - KMeans and OC partitioning clustering algorithms) respectively. K-means clustering algorithm is fast, easy to implement, converges to local optima almost surely but there is one drawback behind k-means clustering algorithm that is it is easily effected by noise. On the other hand, O-Cluster algorithm creates a hierarchical grid-based clustering model, that is, it creates axisparallel (orthogonal) partitions in the input attribute space. The resulting hierarchical structure represents an irregular grid that tessellates the attribute space into clusters. O-Cluster separates areas of high density by placing cutting planes through areas of low density. O-Cluster needs multi-model histograms (peaks and valleys). If an area has projections with uniform or monotonically changing density, then faces difficulty in partition. In this research paper we proposes a new algorithm for merging of clusters and proposed algorithm name is “DRID” which follows a common strategy in two different- different environments and then performance analysis of a distance based algorithm and Grid based algorithm and compare the results. The reason for merging of clusters is to improve the quality of clusters, reduce noise problem and increase the performance and reduce the execution time.