Text Summarization Using Document and Sentence Clustering

Abstract

Text documents have important information and it will be very large in size. Getting the relevant information from the text document is very much challenging criteria in the field of information retrieval. This can be done using the text summarization method. A text document is compressed using a summarizing system to produce a new form that conveys the core idea of the content it contains. The issue of information overload demands access to reliable and properly crafted summaries. Users can quickly find the information they need using data minimization. Saving the time and effort from browsing through the entire collection of documents is main advantage of text summarization. The proposed system is focused on an extractive technique of text summarization using a text clustering and word-graph approach. The proposed System uses the term Frequency, Inverse Document Frequency (TFIDF), Jaccard similarity and Euclidian distance which are important techniques for clustering the text. This hybrid approach deals with the novel method for comprising of document and sentence clustering.