Automatic Text Summarization of Article (NEWS) Using Lexical Chains and WordNet

Abstract

Selection of important information or extracting the same from the original text of large size and present that data in the form of a smaller summaries for easy reading is called as Text Summarization. This process of rephrasing is where we get the shorter version of a text document. As such the Summarizer gives the summary of the News. With the help of few algorithms (like Position of the sentence / phrases, Similarity between the sentences in the main body and the title, Semantics, etc) we can create a Summarizer. Text Summarization has now become the need for numerous applications, for instance, market review for analysts, search engine for phones or PCs, business analysis for those who does business. Outline picks up the necessary data in less time. There are two significant methodologies for a synopsis (Extractive and Abstractive outline) which are talked about in detail later. The procedure conveyed for outline ranges from structured to linguistic. In this paper, we propose a system where we centre around the issue to distinguish the most significant piece of the record and produce an intelligent synopsis for them. In our method, we don't require total semantic interpretation for the substance present, rather, we just make a synopsis utilizing a model of point development in the substance shaped from lexical chains. we used NLP, the WordNet, and Lexical Chains and present a progressed and successful computation to deliver a Summary of the Text.