A Comparative Study of Support Vector Machine and Naive Bayes Classifier for Sentiment Analysis on Amazon Product Reviews

This paper represents a comparison between two machine learning approaches for analyzing the sentiment of the customers' reviews on Amazon products. Eventually, reviews of a product help the customers to understand the product quality. Incorporating multiple product review factors, including product quality, content, time of the review related to product durability, and historically older positive customer reviews will affect product rankings accordingly. Conversely, the manual approach on a large scale comment is time-consuming leading to an inefficient and unproductive way. In this era of artificial intelligence, machine learning is the most convenient way to train the neural network. So it would be much easier to go through thousands of comments if a model were adopted to polarize those reviews and learn from them. In this research work, firstly, the sentiment of the consumer has been analyzed by the Naive Bayes classifier. Meanwhile, the support vector machine(SVM) has classified the sentiments of the users in binary categories. Hence, the data has been passed through the network model after the preprocessing method named term frequency(TF) and inverse document frequency(IDF) lead to evaluate the feature. To sum up, the goal of this research is to find comparatively better machine learning approaches among SVM and Naive Bayes classifier based on statistical measurement.