A semi-automated review classification system based on supervised machine learning

The field of opinion mining is expanding rapidly with the widespread use of internet for e-commerce and social interaction. One of the interesting use of opinion mining is in the field of online producer-consumer industry. The primary goal of the work presented in this paper is to perform a semi-automated sentiment classification on online product reviews for product evaluation using machine learning. We also aim to induce simplicity in sentiment classification; by using a method called Dual Sentiment Analysis, we relegate the need of using complex human annotations or very high end linguistic tools to solve the polarity shift problem in opinion classification. We also propose use of a pseudo-opposites dictionary based on our training corpus which is domain consistent with the training dataset.