Support Vector Machine

A support vector machine is a supervised learning algorithm that separates classes by finding a decision boundary with the largest practical margin. It matters because model performance depends on how data, training, validation, and evaluation are designed, not only on which algorithm is selected.