Sentiment Classification of Drug Reviews Using Machine Learning Techniques

Abstract

Sentiment analysis extracts people's feelings and attitudes about a certain subject. It has recently received a lot of interest in a variety of applications. In general, the sentiment analysis of healthcare, especially of drug experiences of users, might give substantial importance to how to enhance public health and make sound judgments. In this paper, new approaches have been developed that are based on patient reviews to predict sentiment to improve data analysis. Then, use Term Frequency-Inverse Document Frequency (TF-IDF) to extract the features. The experimental findings show that the Random Forest Classifier (RFC) beats all results of other existing models from the literature in terms of Precision, Recall, F1-Score, and Accuracy of 93 % accuracy.