Analysis of Various Machine Learning Techniques used for Automatic Text Summarization

The growing availability of documents has necessitated extensive research in the field of Natural Language Processing (NLP) for Automatic Text Summarization (ATS). Researchers have been striving to develop approaches for creating summaries since the 1950s, such that the machine-generated summary matches the human-made summary. Because manual text summarizing is a time-consuming and inherently tedious activity, automating it is expanding in popularity and so serves as a powerful motivator for academic study. This paper presents taxonomy for ATS and discusses about various approaches that are used for summarization of text. The paper outlines a basic structure of a generic framework for ATS. Various datasets that are frequently used in this domain is also provided. Comparative analysis of various prominent text summarizers on the basis of technique used, language of text, document source and summarization domain is also provided. Various issues faced by the researchers and future prospects are also discussed.