Title: Aspect-based summarisation in the big data environment

Authors: K. Krishnakumari; E. Sivasankar

Addresses: Department of Computer Science and Engineering, A.V.C. College of Engineering, Mannampandal, Mayiladuthurai 609 305, TamilNadu, India ' Department of Computer Science and Engineering, National Institute of Technology, Tiruchirappalli, Tiruchirappalli 609 015, TamilNadu, India

Abstract: Due to the large amounts of information available, it is difficult for customers to select a superior product. With the large volume of information, it is difficult for customers to assess all of the reviews. Sentiment analysis plays an active role in extracting and identifying the opinion of the customer who purchased the product. Sentiment summarisation helps the customer to buy the best product based on its features and values. Our technique involves aspect-based sentiment analysis followed by summarisation. To handle large datasets, we propose a parallel approach using the Hadoop cluster to extract features and opinions. By referring to an online sentiment dictionary and interaction information (IIn) method, the sentiments are predicted and then summarised using clustering. After classifying each opinion word, our summarisation system generates a short summary of the product based on several features. This makes the customer feel comfortable and improves the competitive intelligence.

Keywords: sentiment summarisation; opinion; aspects; Hadoop; MapReduce; big data.

DOI: 10.1504/IJAIP.2023.130816

International Journal of Advanced Intelligence Paradigms, 2023 Vol.25 No.1/2, pp.68 - 83

Received: 08 Jun 2017
Accepted: 17 Feb 2018

Published online: 11 May 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article