The drugs sentiment analysis has grown significantly in importance in the modern day since categorizing medications based on their efficacy by examining user evaluations can help possible future customers in learning more and making better selections about certain drugs. Using social media, blogs, and other sources, sentiment analysis is a popular approach for obtaining opinions from a large number of people. In addition to a lack of accessibility, many people started using medications on their own without the necessary consultations it increased the severity of the health issue. Numerous fields have found applications for machine learning, and creative work for automation is increasing. The information intends to show a medication recommendation method that could significantly decrease the need for specialists. Implementing machine learning methods, such as light gradient boosting machine (LGBM) K-nearest neighbours (KNN), CAT (computer-aided translation) boost, decision tree (DT), and, in order to assess the information and create a model that conducts sentiment classification. Measurements for precision, accuracy, recall, and F1 score were utilized to evaluate the attitudes that were shown. As a consequence, the outcomes demonstrate that the classifier outperforms other approaches.