• Bhargav Ayare
• Amar Singh
• Vanshika Shetty
• Jaya Jesvani

Given the abundance of medications available for the same disease, selecting the best one based on patient evaluations can be difficult. In order to suggest medications with the most positive evaluations, this research proposes a revolutionary medicine recommendation system that makes use of sentiment analysis and machine learning. The UCL Drug Review Dataset, which includes more than 160,000 reviews of different drugs and medical problems, is used by our system. For sentiment classification, we used the Passive Aggressive Classifier and Text Frequency-Inverse Document Frequency (TFIDF), with accuracies of 75.8% and 79%, respectively. In contrast to current algorithms that only classify reviews as favorable or bad, our model uses the sentiment scores to directly recommend drugs. We used methods like Word2Vec for feature representation in our exploratory data analysis, and we preprocessed the data by removing stop words and misspellings to improve its quality. Future research has not yet been detailed, although it may concentrate on adding user-specific variables to suggestions and growing the dataset to include more illnesses and medications.

Drug Recommendation, Sentiment Analysis, EDA, Machine Learning, TF-IDF, Passive Aggressive Classifier, Word2Vec, Data Preprocessing, Feature Representation

"Drug Recommendation System based on Sentiment Analysis using Machine Learning", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 4, page no.n770-n774, April-2024, Available :http://www.jetir.org/papers/JETIR2404D95.pdf

"Drug Recommendation System based on Sentiment Analysis using Machine Learning", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 4, page no. ppn770-n774, April-2024, Available at : http://www.jetir.org/papers/JETIR2404D95.pdf