A Study on the Grid Based Clustering Algorithms

                                                                    Hussian F. Sindi

Computer Science Department, Faculty of Science

King Abdulaziz Univ. Jeddah, Saudi Arabia

dr_sindi@awalnet.net.sa

          Abstract

There are several types for clustering algorithms. The main four are: Partitioning method, hierarchical method, grid-based method and density based method. A clustering algorithm can be employed either as a stand alone tool or as a preprocessing step for other algorithms which operate in the detected clusters. In order to choose a clustering algorithm that is suitable for a particular application, many factors have to be considered. These include: (a) Application goal, (b) Tradeoff between quality and speed, (c) Characteristics of the data (The types of data attributes, dimensionality, and Amount of noise in data). This paper discusses the characteristics of the recent grid-based Clustering algorithms, namely, STING, Wave Cluster, CLIQUE and DENCLUE.

Keywords:

            Clustering Analysis, Spatial Data mining, Knowledge discovery, databases