Here we are making a predictive system to measure the sentiment of each review tweet, whether it is 1 (Positive Sentiment) or 0 (Negative Sentiment). In this work, seven difference machine learning models - LGBM Classifier, XGBooost Classifier, CatBoost Classifier, Random Forest Classifier, Gradient Boosting Classifier, K-Nearest Neighbors and Logisitc Regression are used.
LGBM Classifier gives the best accuracy of 90.78
Dataset is present in UCL Machine Learning Repository
Link: https://archive.ics.uci.edu/dataset/462/drug+review+dataset+drugs+com