Machine Learning Model for Adverse Drug Reaction Detection Based on Naive Bayes and XGBoost Algorithm

Adverse drug effects, commonly referred to as adverse drug reactions (ADRs), represent undesirable and unintended responses to medications or pharmaceutical products when used at recommended doses for therapeutic purposes. These effects can range from mild, tolerable symptoms to severe, life-threatening conditions and can manifest in various ways, affecting different organ systems within the human body. ADE analysis plays a pivotal role in prioritizing patient safety. By meticulously examining the relationship between drug administration and patient responses, healthcare providers can tailor medications to individual profiles, minimizing risks of adverse reactions. This ensures a patient-centric approach to treatment, where prescriptions are finely tuned to maximize efficacy while minimizing potential harm. This research aims to address this challenge by developing a machine learning system utilizing the Naive Bayes and XGBoost algorithms to enhance the categorization of drugs with adverse effects, ultimately contributing to improved patient safety and healthcare decision-making. In our approach, we made a system that detects ADR to effectively combine and collate patient medical history and drug information to detect if a patient would suffer adverse effects or reaction after taking the medication in its correct expert prescribed dose. The XGBoost algorithm gave a 75% accuracy score while Naive Bayes algorithm gave a score of 99%.