Machine Learning Predictions of Drug-Drug Interactions: A Comprehensive Analysis
G. Supriya Reddy, J. V. S. Chandana, P. Roshan Ali, V. V. Rajesham · 2025
Machine learning (ML) has emerged as a powerful tool in the prediction and analysis of drug-drug interactions (DDIs), offering the potential to enhance drug safety and optimize therapeutic outcomes. This study presents a comprehensive analysis of ML predictions of DDIs, leveraging diverse datasets and advanced computational algorithms. We systematically review and compare the performance of various ML models in predicting DDIs, considering factors such as sensitivity, specificity, and area under the receiver operating characteristic curve. The study also explores the impact of data sources, feature selection methods, and model interpretability on prediction accuracy. Our findings contribute to the ongoing efforts in harnessing ML for DDI prediction, providing insights into the strengths, limitations, and future directions of this evolving field.