Real-time Sentiment Analysis On E-Commerce Application

Abstract

Opinion mining is one of the most important tasks of natural language processing, which is also known as sentiment analysis, used to identify about what people have an impression about their services and products in social media platforms. To improve marketing strategies using product reviews, an effective method should be used for predicting the sentiment polarity. In this research article, a Machine learning technique called Support Vector Machine (SVM) is used to design a model and this model has been implemented on an E-commerce application. The data used in this study are online product reviews which are collected from Amazon.com. The experiments of sentiment analysis are performed for two levels of categorization: review level and sentence level. The main focus of this paper is to present a real-time sentimental analysis on the product reviews of e-commerce application so that the user experience can be enhanced.