Which term refers to the data points most critical to defining the decision boundary in SVMs?

Which term refers to the data points most critical to defining the decision boundary in SVMs?

• A. Boundary
• B. Margin
• C. Support Vectors
• D. Centroids

Answer: (C)

In SVMs, the decision boundary (the separating hyperplane) is determined by the data points that lie closest to it. These are the support vectors: the samples that sit on the edge of the margin (or within it in soft-margin SVM). They carry the nonzero influence in the model, meaning the hyperplane’s position and orientation are set by these few points. Points farther from the boundary don’t affect where the boundary ends up once the margin is defined, so they don’t contribute to shaping the decision boundary. The term used for these critical points is support vectors, which distinguishes them from the margin (the distance to the boundary) or from centroids (class means in clustering) or from the boundary itself.

In SVMs, the decision boundary (the separating hyperplane) is determined by the data points that lie closest to it. These are the support vectors: the samples that sit on the edge of the margin (or within it in soft-margin SVM). They carry the nonzero influence in the model, meaning the hyperplane’s position and orientation are set by these few points. Points farther from the boundary don’t affect where the boundary ends up once the margin is defined, so they don’t contribute to shaping the decision boundary. The term used for these critical points is support vectors, which distinguishes them from the margin (the distance to the boundary) or from centroids (class means in clustering) or from the boundary itself.