Drug Target Interaction Prediction Using Support Vector Machine (SVM)

Support vector machine (SVM), a classifier based on machine learning, has also been utilized. The training and evaluation of machine learning was conducted using data from a drug bank. The absence of negative DTI to train on is the greatest obstacle in using machine learning for this purpose. Despite the vast disparity in computing power, the support vector machine (SVM) obtained a superior area under the ROC curve (AUC) of 0.753 0.006 to the most advanced network-based method's 0.886 0.010. After extensive testing, we determined that SVM provided the maximum level of accuracy, 93.76 percent. This was unexpected and may indicate the existence of previously unknown DDI varieties or the maturation of scientific methodologies for studying DDIs. It could be used to characterize several DDI types that were not discovered until advanced processing methods or instruments, such as high-throughput screening, were developed.