Drug Quality Classification Using Sentiment Analysis of Drug Reviews

Since the pandemic emerged, there has been a shortage in remote access to medical resources, a lack of medical experts and healthcare professionals, an absence of adequate instruments and medicines etc. As a result, many people have passed away. As there is absence of resources, people started off following medical recommendations without sufficient consultation, giving rise to their wellbeing to decline should have been. In recent years, machine learning has found several useful uses. The main objective for the study was to give a drug review prediction which would help to decrease the number of health care specialists and authority needed. In this investigation, we are developing a drug review system that analyses sick persons feedback to forecast patient’s comments using various vectorization algorithms, such as the Count Vectorizer, TF-IDF, Bag of Words, and handy analysis of characteristics that can aid in identifying the most effective course of treatment for a condition using several classification methods. The predicted results were assessed using the precision, accuracy, AUC, and F1 scores. The results demonstrate that the Random Forest Classifier surpasses every other model with an accuracy rate of 94%.