Text Summarization using Textrank and Lexrank through Latent Semantic analysis

Abstract

This paper's goal is to investigate the applicability of three distinct text summary algorithms to the realm of blogs, with the ultimate goal of identifying the most precise method for indicating a highly trustworthy automatic summarization tool. The most important component of summarizing is selecting a sample that is reflective of the whole and yet stands on its own as a complete entity. Multiple industries now make use of summarization tools. Automatic document summarizing seeks to provide an informative abstract of a text by extracting its most relevant phrases. Document Concisely summarizing online material is a business that's increasingly in demand now. Blogs have a crucial role in the development and dissemination of public sentiment. Blogs are a popular and powerful medium of online expression, with an estimated 152 million sites now online. Because of this, analyzing blog comments is crucial for describing patterns in consumer spending, evaluations in the fields of social science, health, and agriculture, etc. It's possible that studies in fields as diverse as psychology, anthropology, economics, political science, etc. might benefit from this information on consumer expenditure. The tools used to compile these summaries need to be in step with the analysis's needs. There are several tenors, and each one calls for a unique algorithm based on a specific set of mathematics and computational principles. It is thus necessary to examine the algorithms themselves in order to explain the best method for examining blog-based views.