An Empirical Analysis of Similarity based Single Document Summarization

摘要

In current days, there is an enormous amount of information available in various resources like www, e-books, tweets, several articles, and social media posts. Human beings have very little time available to them, whereas they must obtain the necessary information from these resources due to public and business requirements. Text Summarization is one of the solutions to this problem. It is a process of shortening the length of text present in the document computationally and permit to obtain the most vital information in a very less amount of time. Therefore, a similarity-based framework is suggested to summarize a document concerning various similarity measures. Document Understanding Conference (DUC) dataset is considered for empirical validation of the model.