Ecommerce Product Rating System Based on Senti-Lexicon Analysis
Bahar Uddin Mahmud1, Shib Shankar Bose2, Md. Mujibur Rahman Majumder3, Mohammad Shamsul Arefin4, Afsana Sharmin5

1Bahar Uddin Mahmud*, Computer Science & Engineering, Feni University, Feni, Bangladesh, India.
2Shib Shankar Bose, Computer Science & Engineering, NIT-Silchar, Silchar, India.
3Md.Mujibur Rahman Majumder, Computer Science & Engineering, Feni University, Feni, Bangladesh, India.
4Dr. Mohammad Shamsul Arefin, Computer Science & Engineering, Chittagong, Bangladesh, India.
5Afsana Sharmin, Computer Science & Engineering, Chittagong University of Science & Technology, Chittagong, Bangladesh, India.
Manuscript received on May 07, 2020. | Revised Manuscript received on May 19, 2020. | Manuscript published on June 10, 2020. | PP: 369-373 | Volume-9 Issue-8, June 2020. | Retrieval Number: 100.1/ijitee.H6437069820 | DOI: 10.35940/ijitee.H6437.069820
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (

Abstract: E-commerce is one of the popular systems for buying and selling the products. In comment section of products that they have purchased, customer express their opinion based on the quality of product, the attitude of vendor, the delivery of product etc. This information acts as a reference for the new customers, whether they have bought the product or not. To evaluate the users’ comments, sentiment analysis is played important roles where this approach not only focuses on the product itself but also the features of product itself. In this work, We have calculated the score /rating of user’s sentiment for Amazon products i.e. Mobile phone; by taking the comments from the review section of product which is implied by some words or phrases, are very significant and meaningful to express users’ opinion. This approach performs sentiment analysis using lexicon based approach with the help of Natural Language Toolkit (NLTK) and compare the result with the Amazon’s own product rating. The experimental results prove the effectiveness of the approach. 
Keywords: Product rating, E-commerce, Sentiment analysis, Lexicon, Polarity-text, NLTK.
Scope of the Article: E-commerce