International Journal of Data Mining and Emerging Technologies

  • Year: 2015
  • Volume: 5
  • Issue: 2

Testing Methodologies and Exploring Challenges and Issues in Text Summarisation

1Research Scholar, C.U. Shah University, Wadhwan, Surendranagar, Gujarat, India

2Associate Professor, Adani Institute of Infrastructure Engineering, Ahmedabad, Gujarat, India

*(Corresponding author) Email id: darshnabpatel@gmail.com;

**drhiteshrc1@gmail.com

null

Abstract

Text summarisation is a text that is derived from one or more input text sources, which preserves-an important portion of the information from the original text and significantly less than half of the original text. As the on-line information is increasing day-by-day on internet, users have to make more efforts to find the suitable information from online documents. This situation required research in the area of text summarisation within the Natural Language Processing (NLP) area. Early research started to summarise single document but as the amount of information increased, multi document summariser initiated, which gives summary from set of documents with the same topics or events. This survey studies mainly extractive, abstractive and hybrid approaches in text summarisation. This study intends to scrutinise some of the most relevant approaches for single- and multiple- document summarisations.

Keywords

Automatic text summarisation, Extractive summarisation, Abstractive summarisation, Natural language processing, Evaluation measures, Statistic approaches, Linguistics approaches