Extractive Text Summarization Using Deep Learning

Abstract

An approach for generating short and precise summaries for long text documents is proposed. Lately, the size of information on the internet is increasing. It has become tough for the users to dig into the loads of information to analyze it and draw conclusions. Text summarization solves this problem by generating a summary, selecting sentences which are most important from the document without losing the information. In this work, an approach for Extractive text summarization is designed and implemented for single document summarization. It uses a combination of Restricted Boltzmann Machine and Fuzzy Logic to select important sentences from the text still keeping the summary meaningful and lossless. The text documents used for summarization are in English language. Various sentence and word level features are used to provide meaningful sentences. Two summaries for each document are generated using Restricted Boltzmann Machine and Fuzzy logic. Both summaries are then combined and processed using a set of operations to get the final summary of the document. The results show that the designed approach overcomes the problem of text overloading by generating an effective summary.