No Detail Found!

Prediction of product rating based on polarized reviews using supervised machine learning

Author(s)
Raheel Ahmad Khan Department of Computer Science, NFC Institute of Engineering and Technology Multan,Punjab,Pakistan
Abdul Mannan Department of Biomedical Engineering, NFC Institute of Engineering and Technology Multan, Punjab, Pakistan; Department of Electrical Engineering, NFC Institute of Engineering and Technology Multan, Punjab, Pakistan
Naeem Aslam Department of Computer Science, NFC Institute of Engineering and Technology Multan,Punjab,Pakistan
Abstract
E-commerce websites facilitate customers to leave their experiences in the form of textual reviews for a variety of products. Recently, online reviews have played significant influencing role in customers' decision for purchasing. The reviews have information and first hand experience about products' quality for customers. Free-text sections are frequently found on online review web pages in addition to star-level reviewing options. But on many web pages, we find only former option. Therefore, there is a need to convert the textual reviews to star-level numerical rating on the basis of the information they contain because business and data analysts prefer numerical rating over a textual review. Therefore, automatic conversion of online text-based reviews into numerical rating has recently been emerged as an active field of research in machine and deep learning. All existing researches in rating prediction have not utilized negative or negation words which play an important role in sentiment expression. In this research, we have used lexicon based method to give equal importance to negative or negation words which improves the rating prediction. This paper presents a supervised machine and deep learning based solution to transform text-based reviews to star-level numerical representation by exploiting polarization detected on the basis of lexical analysis. Experiments were conducted on famous Amazon datasets under different choices of regression and classification techniques. Experimental results have indicated that the use of polarized reviews can significantly improve the rating prediction.
Publication Details
Page(s) 1-9
DOI DOI not available
Published Journal: VFAST Transactions on Software Engineering, Volume: 10, Issue: 4, Year: 2022

Request the full text from PASTIC

Keywords
sentiment analysis, Rating prediction, review rating prediction, review polarization
0

Citations

0

Downloads

84

Views