An Automatic Text Summarization on Naive Bayes Classifier Using Latent Semantic Analysis

Abstract

Currently, huge information is available on Internet, but it is difficult to find the relevant information at a fast and efficient rate. Large collection of textual data is available on the Internet. A very competent system is required to find the most appropriate information from the corpus. Automatic text summarization converts a large document into a shorter precise version. It selects the significant part of the text and builds a comprehensive summary that represents the main content of the given document. Text summarization extracts sentences based on the calculation of the score and rank from the document. In this paper, the model that we have developed uses latent semantic analysis technique and chooses sentences based on a specific threshold given by the system. Further, using Naive Bayes approach of machine learning, the model trains the classifier and predicts the summary that is built on the basis of calculation of singular-value decomposition (SVD). Before training the model, it selects two important concepts of SVD—feature ranking and recursive feature elimination. This paper focuses on extractive text summarization using machine learning, statistical techniques, and latent semantic analysis.