Sentimental analysis of amazon customers based on their review comments

Abstract

Amazon being one of the leading technology giants in the public cloud market is very popular for the services such as amazon shopping and prime video for online shopping and online video streaming. Now a day's customers buy products based on the product reviews given by already purchased customers. If the reviews are good, then that product rating is high. Sentimental analysis is a fastest growing research area where text mining plays a major role in analysing the customer reviews. There are many methods currently existing on sentimental analysis of amazon customers based on their review comments such as using machine learning algorithms and deep learning algorithms. This article proposes a hybrid approach by combining technologies such as Natural language processing, Machine Learning and Deep Learning methods to analyse the customer reviews and ratings. Based on the negative, positive comments and ratings the sentiment of customer will be analysed. The cleaned dataset is pre-processed and sentiment analysis is done using Machine Learning and NLP. The work is proposed to perform sentiment analysis on the data extracted from the pre-processed data of the customer review and ratings and best classifier algorithms such as SVM., NB with natural processing language. The positive and negative classification in done depending on the customer reviews and ratings by the ML techniques. The novelty of the work is that combination of techniques such as NLP., ML and DL are applied.