Supervised Sentiment Analysis on Amazon Product Reviews: A survey

Abstract

Sentiment Analysis (SA), which is also known as Opinion Mining, is a hot-fastest growing research area, making it challenging to follow all the activities in such areas. It intends to study people's thoughts, feelings, and attitudes about topics, events, issues, entities, individuals, and their attributes in social media (e.g., social networking sites, forums, blogs, etc.) expressed by either text reviews or comments. Amazon is an example of the world's largest online retailer that allows its customers to rate its products and freely write reviews. Analyzing these reviews into positive or negative; will assist customers' decision making, which varies from purchasing a product like a camera, mobile phone, etc., to writing a review about movies and making investments - all of these decisions will have a significant impact on the daily life. Sentiment analysis draws the attention of both scientific and market research in Natural Language Processing and Machine Learning fields. In general, the machine learning approach consists of supervised and unsupervised algorithms. In this research study, a detailed typical workflow process often adopted by the researchers is presented. Moreover, traditional supervised machine learning classification techniques have been investigated on various categories of Amazon product reviews to find the best method that provides a reliable result of sentiment analysis and assists future research in this newly emerging area.