Abstract—Innovative methods that are user friendly and efficient are needed for retrieval of textual information available on the World Wide Web. The self-organizing map (SOM) is one of the most widely used neural network algorithms. SOM can be used to identify clusters of documents with similar context and content. In this paper, we explore and visualize the Self Organizing Map and discuss how to classify text documents. The paper also portrays the capabilities of SOM in text classification. We also discuss about experiments done using 20 news group dataset.
Index Terms—Self-organizing map, stop words, term-document, neurons.
Manuscript received Tue, July 7, 2009. This work was is carried out as one of a part of the research work titled “An Intelligent Agent for Efficient Internet Searching” by B. H. Chandra shekar. B. H. Chandra Shekar, is working as Lecturer in the Department of Master of Computer Applications, R. V. College of Engineering, Mysore Raod, Bangalore 560059, India. (e-mail: firstname.lastname@example.org)
Dr. G. Shobha is the Director of Master of Computer Applications, R. V. College of Engineering, Mysore Road, Bangalaore 560059, India. (e-mail: email@example.com)
Cite: B. H. ChandraShekar and G. Shoba, "Classification Of Documents Using Kohonen’s Self-Organizing Map," International Journal of Computer Theory and Engineering vol. 1, no. 5, pp. 610-613, 2009.