Identifying Biased Reviews: An Analysis on Amazon Electronic Products

Identifying Biased Reviews: An Analysis on Amazon Electronic Products

Md. Niaz Imtiaz, Md. Toukir Ahmed, Md. Rakib Hasan
Copyright: © 2022 |Pages: 10
DOI: 10.4018/IJSI.297991
OnDemand:
(Individual Articles)
Available
$37.50
Current Special Offers
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

Online reviews play a significant role in our community contributing to the prediction of the marketing situation, making industries modifying their advertising policies. Many consumers choose online reviews for making the decision to buy a specific product. In recent years, product sellers provide some lucrative offers to write biased reviews which are usually very positive that increases the rating of the products significantly. So it is very important to detect biased reviews for online shopping to help the consumers in their decision making to buy proper products. In this work, a new method has been developed for detecting those biased reviews generated on some products at Amazon. At first online reviews of Amazon product like- Fire Tablet, Alkaline Batteries, etc. are collected. Then sentiment analysis is introduced for calculating the sentiment score of the text reviews with the help of natural language processing. Naïve-Bayes-Analyzer model and TextBlob library are used to calculate the sentiment scores. Finally, statistical measurements are used to detect biased reviews.
Article Preview
Top

Background

Different researchers have been worked on biased reviews detection. In this section we focused on some sentiment analysis technique with NLP to find biased reviews from reference section.

In the study, authors established a procedure to carry out the outcome revealing, online reviews are statically biased. Authors introduced string searching algorithm for analyzing the reviews and then compared the non-biased reviews with the statistical average score for biased reviews to determine if there is a difference in population means. Furthermore, the author used statistical modeling to determine the biased reviews are statistically significantly different or not. The results produced from the data analysis confirmed that biased reviews are statistically significantly higher than non-incentivized reviews.

Complete Article List

Search this Journal:
Reset
Volume 12: 1 Issue (2024)
Volume 11: 1 Issue (2023)
Volume 10: 4 Issues (2022): 2 Released, 2 Forthcoming
Volume 9: 4 Issues (2021)
Volume 8: 4 Issues (2020)
Volume 7: 4 Issues (2019)
Volume 6: 4 Issues (2018)
Volume 5: 4 Issues (2017)
Volume 4: 4 Issues (2016)
Volume 3: 4 Issues (2015)
Volume 2: 4 Issues (2014)
Volume 1: 4 Issues (2013)
View Complete Journal Contents Listing