Open AccessAspect Based Sentiment Analysis for E-Commerce Websites with Visualization through Machine Learning Algorithm
Author(s) -
Ms. Nandini N S,
M. P.
Publication year - 2020
Publication title -
international journal of innovative technology and exploring engineering
Language(s) - English
Resource type - Journals
ISSN - 2278-3075
DOI - 10.35940/ijitee.e2838.039520
Subject(s) - computer science , sentiment analysis , e commerce , product (mathematics) , pace , aggregate (composite) , machine learning , artificial intelligence , representation (politics) , information retrieval , world wide web , materials science , geometry , mathematics , geodesy , politics , political science , law , composite material , geography
E-commerce is evolving at a rapid pace that new doors have been opened for the people to express their emotions towards the products. The opinions of the customers plays an important role in the e-commerce sites. It is practically a tedious job to analyze the opinions of users and form a pros and cons for respective products. This paper develops a solution through machine learning algorithms by pre-processing the reviews based on features of mobile products. This mainly focus on aspect level of opinions which uses SentiWordNet, Natural Language Processing and aggregate scores for analyzing the text reviews. The experimental results provide the visual representation of products which provide better understanding of product reviews rather than reading through long textual reviews which includes strengths and weakness of the product using Naive Bayes algorithm. This results also helps the e-commerce vendors to overcome the weakness of the products and meet the customer expectations.