Feature Based Sentiment Analysis of Mobile Product Reviews using Machine Learning Techniques

Abstract

With the advent of Web technology and its rapid growth led many people to express their views or opinions on the web related to the products, the services, the events or any social issues.Due to the substantial extent of data which is increasing day by day on the web, it is obvious that the reviews provided by the people on the e-commerce sites is also huge and such reviews are in unstructured form.Hence, it is challenging for the online users, customers and manufacturers to make a proper decision about opinions on these reviews.During the last two decades, many researchers including corporation are working consistently on sentiment analysis, also recognized as, opinion analysis or mining, to derive the opinions expressed by various end users as reviews.In this paper, we study the fine-grained investigation in identifying the sentiments (or opinions) expressed on various aspects of the entity considering the aspects as explicit one over various brands of mobile product reviews and classify these opinions based on some machine learning algorithms.Exploratory outcomes on the different datasets manifest the promising results with respect to the accuracy of classifying the opinions.