A Unified Framework for Recommend Medicine Review Categories Through Bi-LSTM and Supervised Learning

The number of users and daily tweets on Twitter are significant. The medicine review categories system is designed to leverage the vast amount of patient-generated data to create a reliable and effective medication recommendation framework. The proposed system employs Deep Learning and Machine Learning techniques, specifically utilizing a deep learning technology known as Bi-Directional Long Short-Term Memory (Bi-LSTM) and machine learning algorithms Decision Tree, Random Forest, and Naive Bayes to evaluate textual data effectively. The primary objective of our system is to identify appropriate categories of input text /review based on patient reviews and defined categories in the dataset. In the dataset, we found two columns where one is a review and another is categories like treatment and medication, prevention, symptoms causes, news, and others where we will find the exact categories of our input review. We evaluate the system's performance using various metrics, including Fl Score, precision, recall, and accuracy, ensuring its robustness and efficacy. Bi LSTM models provided an accuracy of 95%, the Decision Tree models provided an accuracy of 94 %., the Random Forest models resulted in an accuracy of 92%., and the Naive Bayes models provided an accuracy of 84%. This comprehensive approach aims to provide medical experts with a reliable tool for predicting and ultimately contributing to more informed and personalized healthcare decisions.