Approaches for Automatic Text Summarization: A Brief Survey

Abstract

Propelled by exponential growth in the latest technological trends and innovations, data is very crucial to this century as oil was to the previous one. Now a day’s world is parachuted by the assembling as well as propagation of vast amount of digital data. Due to the huge amount of information present in todays digital era, There is crucial need to provide the efficient and effective method the extraction of information quickly.So, machine learning algorithms needs to automatically reducing long text sentences to build a coherent and fluent summary by maintaining its overall content and purpose of generated summary. It is very tricky and lengthy process for anyone to manually extract meaningful summary of large document so here comes need of automatic text summarization. Text summarization has been and continues to be a hot research topic in the data science arena. There are two main approaches of Automatic Text summarization as a  Extractive Text Summarization and Abstractive Text Summarization. This paper gives review on a comprehensive literature work on abstraction based text summarization techniques. Abstraction based  techniques are broadly cotegorized into two classes as structure based method and semantic based method. The summarization and interpretation of the different  techniques used for automatic text summarization and their challenges is the main focus of this paper.From the existing literature work,this paper concludes that abstractive text summarization techniques creates most cohesive, coherent, less redundant and meaningful abstract to the greatest extent. Still there are some open research issues which are recognized in abstraction  based text summarization methods,need to be address and resolved in future research work.