Online Product Reviews and Their Impact on Third Party Sellers Using Natural Language Processing

Akash Nellutla, Manoj Hudnurkar, Suhas Ambekar, Abhay Lidbe

OAI: oai:igi-global.com:269445 • DOI: 10.4018/IJBIR.20210101.oa2

The purpose of this paper is to gain insights from the online product reviews of e-commerce sites such as Flipkart and Amazon and analyze its impact on third party sellers. To judge the authenticity of a product, reviews are more useful than ratings, since ratings do not give a complete picture. It is always preferred to consider both the product and seller reviews to have a seamless delivery and defect less product. In this paper, natural processing methods are used to gain insights by considering online reviews of a product. Methods such as sentiment analysis, bag of words model help to understand the impact of online product reviews on the seller's ratings and their performance over some time. The reviews are categorized into positive, negative, and neutral using sentiment analysis. Further, topic modeling is done to find out the topic reviews are majorly referring to. The seller reviews for a specific product after analysis are compared with the overall seller reviews to judge the authenticity. The results of this paper would be beneficial to both the consumers and sellers.

Business Intelligence Business and Management Business Information Systems