Title: Reviewer reliability with sentiment analysis using DECE optimiser

Authors: S.S. Arumugam

Addresses: Department of Computer Science and Engineering, Saveetha School of Engineering, Saveetha Institute of Medical and Technical Sciences, Chennai, India

Abstract: Due to the rapid advancement of internet technology in recent years, online shopping has become facilitated to users for selling and purchasing. The first step is to pre-process the data from Amazon's product recommendations before feeding it to an ensemble classifier made up of large scale pinball twin support vector machine (LSPTSVM), categorical gradient boosting (CGB) and integrated fuzzy decision tree (IFDT) that classifies the outcomes into three categories: good, bad, average. In general, the classifier does not adapt any optimisation models to define the optimum parameter and to achieve the exact classification. This manuscript proposes a chaotic equilibrium optimiser (CEO) technique to optimise the classifier that classifies precisely. The proposed approach is activated in MATLAB. The proposed method reaches a lower mean absolute error 28.32%, 22.82% and 23.56%, a higher mean absolute percentage error 28.62%, 19.82% and 22.36% and a mean square error 25.31%, 20.33% and 14.2% less than the existing systems.

Keywords: categorical gradient boosting; CGB; chaotic equilibrium optimiser; CEO; deep ensemble classifier; integrated fuzzy decision tree; IFDT; large scale pinball twin support vector machine; online product shopping reviews; sentiment analysis; recommendation system.

DOI: 10.1504/IJMC.2025.148005

International Journal of Mobile Communications, 2025 Vol.26 No.2, pp.239 - 255

Accepted: 18 Jun 2024
Published online: 13 Aug 2025 *

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