An Efficient Clustering Algorithm for Text Mining Using Greedy Approach

I. Introduction " Data Mining " involves the integration of concepts from computer science, mathematics, and statistics. It seeks to extract useful information and detect interesting correlation and patterns from any form of data, especially numeric data. Data Mining is most associated with the broader process of Knowledge Discovery in Databases (KDD), " the nontrivial process of identifying valid, novel, potentially useful and ultimately understandable patterns in data " (Fayyad et al., 1996). By analogy, " text mining " as the process that exploits large text collections to obtain valid, potentially useful and ultimately understandable knowledge. It is important to emphasize that getting from a collection of documents to a clustering of the collection, is not merely a single operation, but is more a process in multiple stages. These stages include more traditional information retrieval operations such as crawling, indexing, weighting, filtering etc. Some of these other processes are central to the quality and performance of most clustering algorithms, and it is thus necessary to consider these stages together with a given clustering algorithm to harness its true potential. We will give a brief overview of the clustering process, before we begin our literature study and analysis. We have divided the offline clustering process into the four stages outlined below: