Drug Recommendation System and Side Effects Based on Sentiment Analysis of Drug Reviews Using Machine Learning

This study provides a medicine recommendation system that analyses the problems by applying machine learning and sentiment analysis approaches. Classification algorithms are used to predict sentiment and provide suitable medications based on these assessments. The powerful machine learning (ML) approaches can help to further improve this, which analyses unstructured data sources like user ratings, medical literature and patient reviews. For appropriate drug recommendations with no adverse effects, this study suggests a Priority Weight Allocation Model using User Ratings (PWAUR). Based on the results, which demonstrate 98% accuracy, the Gradient Boosting model using TF-IDF outperforms the other models. By effectively predicting and managing side effects, these systems can significantly improve patient outcomes. Index Terms: drug, sentiment analysis, TF-IDF, NLP, Stemming, Recommender System, Machine Learning, Gradient Boosting and BOW.