Litcius/Paper detail

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

Satvik Garg

2021105 citationsDOIOpen Access PDF

Abstract

In the post-pandemic era, access to timely and accurate medical treatment has become a significant challenge.A growing number of patients are turning to online platforms to self-medicate based on peer-shared drug reviews, often bypassing medical consultations.This study proposes an intelligent drug recommendation system that harnesses sentiment analysis of drug reviews using machine learning (ML) techniques.Utilizing Natural Language Processing (NLP) and vectorization techniques (TF-IDF, bag of words, and Word2Vec), we trained several classifiers, including linear SVC, logistic regression, and random forest, to predict user sentiment toward medications.Among these, the Linear SVC model using TF-IDF vectorization achieved the highest performance, with an accuracy of 93%.The system integrates these predictions to recommend the top drugs for specific medical conditions based on user feedback.Our study demonstrates that sentiment-aware drug recommendations can support better data-driven healthcare decisions.

Topics & Concepts

Recommender systemMachine learningComputer scienceArtificial intelligenceEconomic shortageClassifier (UML)Sentiment analysisFeature (linguistics)Drug repositioningVectorization (mathematics)Health carePrecision medicineDrugHealth informaticsCoronavirus disease 2019 (COVID-19)Topic modelData scienceFeature engineeringInformation retrievalData miningFeature extractionCollaborative filteringNatural language processingStatistical classificationMachine Learning in HealthcareSentiment Analysis and Opinion MiningText and Document Classification Technologies