Text Summarization using different Methods for Deep Learning

Abstract

In the current era of rapid information expansion, text summarization has become vital for comprehending textual material. Physically condensing large textual volumes is challenging for humans, especially considering the vast amount of text content available online. Text summarization is an active field of research that focuses on summarizing large texts into shorter versions while retaining important information. The writing can be categorized as either extractive or abstractive regarding its summary. Extractive summarizing methods function by determining the significance of individual sentences within a text and selecting them to form a summary. This approach relies solely on sentences extracted from the original text. Abstractive summarization methods aim to rephrase significant information. Text summarization can be achieved by many deep learning methodologies, including fuzzy logic, Convolutional Neural Networks (CNN), transformers, neural networks, and reinforcement learning. Over the past three years, there has been a modest shift in the research focus on text summarizing. Current developments aim to increase the efficiency of text summarization and attain optimal accuracy. This study aims to examine the various methods of using deep learning for text summarization and identify the current deep learning developments.