Using lexical chains for efficient Text summarization

The emergence of Internet technology has become a dominant factor in our education, business, and everyday life. In recent years, with

the rapid expansion of Internet capability, it has become difficult for end-users to efficiently access the enormous amount of information provided for their consumption within a limited time. This problem, therefore, required an efficient tool to help manage this vast quantity of information. For this reason, any application that has the ability to summarize information automatically and present results to the end-user in a compressed, yet complete form; would be a good attempt to the solution of this problem.

 

In this paper, our primary goal is to discuss and present an efficient and effective tool that is able to summarize large documents quickly while preserving their content. We investigate a summarization method which uses not only statistical features but also relative and contextual meaning of documents by using the lexical chain, which is a method of capturing the “aboutness” of a document. We present a new algorithm to compute lexical chains in a text with robust and economical knowledge resources: the WordNet thesaurus.

 

In this algorithm, summarization proceeds in four steps: the original text is segmented, lexical chains are constructed, strong chains are identified, and significant sentences are extracted. We show that our method is efficient and tends to provide quality indicative summaries within a short time. We briefly identify unresolved problems and address the future scope and plans of the method.