Automatic Text Summarization using Extractive Techniques and Attribute Tagger Algorithm

Abstract

With the emotional development of the Internet, individuals are overpowered by the enormous measure of
online data, archives and documents. This growing accessibility documents or records have requested
comprehensive exploration in the region of automatic text summarization. Text Summarizer is a method of
shortening long pieces of text. The expectation is to make a cognizant and fluent summary having just the primary
concerns delineated in the document. Automatic Text Summarization techniques are classified in to two, first
one Abstraction based approach and second one Extraction based approach. In this paper we used extractive text
summarization technique along with our novel algorithm attribute tagger to outline the document to present key
information in it. Extraction-based summarization model takes an input that encapsulates some paragraphs and
returns a text summary that represents the outline information or message in the input text. Attribute Tagger
algorithm reduces the work by identifying keywords or important attributes in the input using NER technique.
The output from the Attribute Tagger algorithm is given as input to TextRank and SentenceRank algorithms.
The test results show that our proposed approach can improve the performance compared to sate-of-the-art
summarization approaches.