期刊名称：International Journal of Advances in Soft Computing and Its Applications
印刷版ISSN：2074-8523
出版年度：2016
卷号：8
期号：3
出版社：International Center for Scientific Research and Studies
摘要：The information overload problem has posed great challenge tointernet users to retrieve relevant information accurately for the pastdecades. It is a tedious task for machine to intuitively mimic humanlinguists to summarize documents into meaningful text in abstractivemanner. Quite often, the summarized text lacks cohesion andbecomes difficult to comprehend. The objective of this paper is toinvestigate the proposed Swarm LSA-PSO model performs betterthan alternative methods. In this study, terms matrix was constructedfrom co-occurrence of terms using Bag-of-Words (BOW). The hugedimensions of terms were reduced using Singular ValueDecomposition followed by K-Means PSO clustering for acquiringoptimal number of concepts clusters. These key concepts were usedto identify the main gist in documents for text summarization. Theinput text documents were downloaded from DocumentUnderstanding Conference (DUC) 2002 dataset. The preliminaryresults show that the swarm LSA-PSO model shows promisingresults in context based text summarization using BOW clusteringapproach
关键词：Bag-of-Words; Latent Semantic Analysis; co-occurrence; Text;Clustering; Text Summarization
