Predicting Drug-Drug Interactions Using Machine Learning

Drug-drug interactions (DDIs) can lead to severe adverse effects and pose extreme challenges in clinical practice. Machine learning techniques are employed in this study to predict DDIs with the support of varied classification models. The performance of the Random Forest Classifier, Support Vector Classifier (SVC), and Logistic Regression models has been evaluated in the study. A thorough hyperparameter tuning and evaluation have been performed to determine the model that shows enhanced performance. The result shows that the Logistic Regression model outperforms the other models in F1-score and thus is a useful tool for DDI prediction.