A Survey on Extractive and Abstractive Text Summarization

Abstract

In recent years, the volume of text data from a variety of sources has boomed. This amount
of text is an important source of information and knowledge that needs to be summarised efficiently
for use. Extraction and abstraction are the two separate automated description methods. Extractive
text summarization is based on the principle of sentence salience in order to recognise the most
relevant phrases in a document or document collection. Usually, salience is defined in terms of the
existence of specific important words, or in terms of similarity to a pseudo-sentence centroid. This
paper presents a systematic survey of the new extractive ways to deal with text summarization
delivered in the most recent decade. There are likewise a couple of abstractive and extractive ways to
deal with summarizing text. Another overwhelming challenge in this context is summary evaluation.
Consequently, both natural and outward strategies for summary evaluation are characterized top to
bottom alongside gatherings and workshops on text summary evaluation