Multi document summarization using clustering.

R. C. Balabantaray; D. K. Sahoo; M. Swain; B. Sahoo

Author(s)

R. C. Balabantaray CLIA Lab, Department of Computer Science, International Institute of Information Technology (IIIT), Bhubaneswar, Odisha, India

D. K. Sahoo CLIA Lab, International Institute of Information Technology (IIIT), Bhubaneswar, Odisha, India

M. Swain CLIA Lab, International Institute of Information Technology (IIIT), Bhubaneswar, Odisha, India

B. Sahoo CLIA Lab, International Institute of Information Technology (IIIT), Bhubaneswar, Odisha, India

Abstract

There are two types of situations in which multi-document summarization would be useful: (1) the user is faced with a collection of dis-similar documents and wishes to assess the information landscape contained in the collection, or (2) there is a collection of topically related documents, extracted from a larger more diverse collection as the result of a query, or a topically-cohesive cluster. In this paper we present the development of an automatic multi- document summarizer based on clustering and sentence extraction. Based on reference document provided by the user, similar type of documents is extracted from a group of documents. We create an n x n similarity matrix among the entire sentence in the similar type of documents which represent sentence level similarity in all sentences in all extracted documents. Then we make clusters of similar sentences using Markov clustering principle. Then in each cluster each sentence is assigned three weights 1.chronological weight (Document level) 2.Position weight (position of sentence in document) 3. Sentence weight (Statistical weight based on term weight). Then we extract best sentences from each cluster. We have tested the system with document of different domain documents and the result is satisfactory.

Publication Details

Page(s) 565-571

DOI DOI not available

Published Journal: Journal of Theoretical and Applied Information Technology, Volume: 46, Issue: 2, Year: 2012

Request the full text from PASTIC