Enhanced Drug-Drug Interaction Prediction with Graph Neural Networks and SVM

Predicting drug-drug interactions (DDIs) is crucial for medication safety and efficacy. Traditional methods often face challenges in capturing complex interactions within large-scale drug networks. In this study, we propose an enhanced approach leveraging Graph Neural Networks (GNNs) and Support Vector Machines (SVM) to improve DDI prediction accuracy. Our method integrates graph representation learning with SVM-based classification, effectively capturing intricate relationships between drugs and their interactions. Through extensive experimentation on benchmark datasets, we demonstrate superior performance compared to existing methods, achieving higher prediction accuracy of ${9 7 \%}$ and f1 measure of ${9 8 \%}$. Moreover, our approach offers interpretability, enabling insights into underlying interaction mechanisms. Overall, our study highlights the potential of combining GNNs and SVM for advancing drug interaction prediction, with implications for enhancing medication safety and clinical decision-making.