Document Clustering Using Hybrid Ant Algorithm

In recent years it is required to store/retrieve a huge quantum of documents across network in World Wide Web (WWW) due to the wide spread usage of computers across the globe. This has placed many challenges to the Information Retrieval (IR) system like fetching of relevant documents matching with user’s query, classification of electronic documents etc. Clustering is an unsupervised learning that partitions the available documents into several clusters based on the similarity between the documents. The problem of clustering has become a combinatorial optimization problem in IR system due to the exponential growth in information over WWW. In this paper, a novel Hybrid Ant Algorithm, a blended scheme of Tabu Search and Ant Colony Optimization algorithm has been proposed to form better quality clusters with documents of similar features. The viability of the proposed algorithm is tested over ma few standard benchmark datasets and the numerical experimental results reveal that the proposed algorithm yields promising quality clusters compared to other ones produced by K-means algorithm.