High Impact Factor : 4.396 icon | Submit Manuscript Online icon |

Document Clustering for Effective Information Retrieval System using Genetic Algorithm

Author(s):

V. Raja Manickam , Alagappa University; Dr. A. Nagarajan, Alagappa University

Keywords:

Genetic Algorithm, Information Retrieval, 20Newsgroups Documents, Document Clustering

Abstract

Document clustering is a significant domain of interest in the field of document summarization. K-means clustering is one of the methods used for clustering documents. These methods suffer from issues and challenges like accuracy and time complexity. Information retrieval is the activity of obtaining information resources relevant to an information need from a collection of information resources. The searches of the information retrieval can be based on metadata, full-text or other content-based indexing. In clustering system it can be very useful in web search for grouping search results into closely related sets of documents. It can improve the similarity search on information retrieval. To overcome these limitations a new genetic algorithm based document clustering method have been proposed in this research work. This work also proposed the Boolean operator based information retrieval scheme to find out the particular query raised by the user. This research work the document clustering is performed by 20Newsgroups document dataset. The objective of this research work is to cluster the document by using genetic algorithm and retrieved the user query by based on Boolean operator information retrieval system. This proposed method is implemented and evaluated by various quality measures like confidence value and collective strength. The experimental analysis in this proposed methodology provides the better time complexity, memory utilization and CPU utilization compared with various existing methods.

Other Details

Paper ID: IJSRDV5I60138
Published in: Volume : 5, Issue : 6
Publication Date: 01/09/2017
Page(s): 505-509

Article Preview

Download Article