Supervised and Unsupervised Analysis
Authors/Creators
Description
This study applies supervised and unsupervised machine learning to large-scale pharmaceutical drug reviews to predict patient satisfaction and discover common themes in real-world medication experiences. Models like Naïve Bayes, SVM, and Random Forest were used for classification, with Random Forest performing best. Unsupervised methods such as K-Means and LDA revealed key topics including side effects, benefits, mood changes, and daily-life impact. The work highlights how NLP and ML can convert unstructured patient feedback into actionable healthcare insights.