Deep Learning Based On Different Methods For Text Summary: A Survey

Abstract

Abstract—in today's rapidly growing information age, text summary has become a critical and important instrument for help understanding text information. it is really hard for human beings to physically summarize huge textual documents also there is an abundance of text content available online. text summarization is an active research field that works on compressing large pieces of text into smaller texts that preserve relevant information. text summary classified as extractive or abstractive. methods of extractive summarization working by deciding important text sentences and choosing them as a summary. that method based only on sentences from the source text. methods of abstractive summarization aim to paraphrase important information in a new form like that of humans. text summary can be achieved using different deep learning techniques, such as: fuzzy logic, Convolutional Neural (CNN), transformers, neural network, reinforcement learning, etc. in the past three years, the research trend in text summarization has also undergone a slight change, where new trends have appeared that are trends that lead to enhancement, how to improve the efficiency of text summarization to obtain high accuracy. we have made several attempts in this paper to discuss the various techniques used on the basis of deep learning for text summary in these years and observe the new trends in the field of deep learning.