Event Title

Sentiment Analysis of Business Reviews

Presenter Information

Merisha SubediFollow

Faculty Mentor

Dr. Nary Subramanian

Document Type

Oral Presentation

Date of Publication

April 2021

Abstract

Sentiment analysis is a technique to gauge the overall sentiment of a statement or series of statements that can be used to determine the attitude of a writer towards the topic in question. The application of sentiment analysis ranges from social media monitoring, product analysis, market research to brand reputation management. In our study, we incorporate sentiment analysis to develop a method that provides the sentiment of reviews of a business provided by its customers. Customer feedback is very important for the success of any business and going through each review manually consumes a lot of time and effort. The purpose of our study is to give the summary of all the reviews of a particular business based on their sentiment. The model features percentage-wise distribution between positive, negative, and neutral reviews with a categorization of positive and negative reviews for any business. Moreover, it provides an unbiased overview of any business with good accuracy. The developed model is based on a combination of web scraping, sentiment analysis, and text analysis, besides providing a graphical user interface for specifying the website to get reviews from. The model has been validated against test data and the result shows 83% accuracy in predicting positive sentiment, 77% accuracy in predicting negative sentiment, and 97% accuracy in predicting neutral sentiment. It can be used to provide valuable user feedback to both consumers and producers for making better decisions about their products and services.

Keywords

Sentiment analysis, Web scraping, text analysis

Persistent Identifier

http://hdl.handle.net/10950/3100

This document is currently not available here.

Share

COinS
 
Apr 16th, 12:00 PM Apr 16th, 1:00 PM

Sentiment Analysis of Business Reviews

Sentiment analysis is a technique to gauge the overall sentiment of a statement or series of statements that can be used to determine the attitude of a writer towards the topic in question. The application of sentiment analysis ranges from social media monitoring, product analysis, market research to brand reputation management. In our study, we incorporate sentiment analysis to develop a method that provides the sentiment of reviews of a business provided by its customers. Customer feedback is very important for the success of any business and going through each review manually consumes a lot of time and effort. The purpose of our study is to give the summary of all the reviews of a particular business based on their sentiment. The model features percentage-wise distribution between positive, negative, and neutral reviews with a categorization of positive and negative reviews for any business. Moreover, it provides an unbiased overview of any business with good accuracy. The developed model is based on a combination of web scraping, sentiment analysis, and text analysis, besides providing a graphical user interface for specifying the website to get reviews from. The model has been validated against test data and the result shows 83% accuracy in predicting positive sentiment, 77% accuracy in predicting negative sentiment, and 97% accuracy in predicting neutral sentiment. It can be used to provide valuable user feedback to both consumers and producers for making better decisions about their products and services.