Volume 18, No. 5, 2021
Employing a Similarity-based Clustering approach for effective text summary
Vikas Tripathi , Dibyahash Bordoloi
Abstract
With so much new data being created every day, summarising documents is becoming more important. Document summarising facilitates comprehension of the text content without reading the whole document. Text summary offers a structure for intentionally condensing one or more text files. It is an important technique for discovering similar content in vast text libraries or on the Web. It is also vital to retrieving the data so that the material is interesting to the user. Extractive summarising and abstractive summarization are the two primary techniques for text summarization. Extractive summarization is a technique for generating summaries by selecting phrases from a text source and ranking them according to their importance. The abstractive summarization approach captures the document's key ideas and renders them in abstract form using natural language. Based on these two methodologies, several strategies for summarising have been created. Several approaches only work with certain languages. This article discusses numerous strategies depending on extractive and abstractive summary approaches and the limitations of these methodologies.
Pages: 3117-3125
Keywords: Document Summarization, Extractive approach, Abstractive approach, Text summary, Information Extraction system.