Skip to content
BENIBIX Learn

Support Vector Machine (SVM)

AI📈 Machine LearningClassic ML Algorithms

Also known as SVM, Support Vector Classifier

It splits two groups with a line. The line leaves the widest gap it can.

Think of it like

Think of it like two towns with a road between them. You make the road as wide as you can.

Example

Lucy's model sorts emails into work and personal. The SVM finds the border with the most room on both sides. So new emails near the edge are sorted more safely.

Why it matters

The wide gap makes it hold up well on new data. It still works well on small data sets with many features.

Definition

A support vector machine splits two groups with the widest gap it can find. It looks for the line that stays as far as it can from the nearest points of both groups. Those nearest points are called support vectors, and they alone decide where the line goes. A trick lets it draw curved borders too. It was a top method before deep learning took over.
The dashed lines mark the widest empty street. Only the dots touching it decide where the middle line goes.

Common mistake

Using it on millions of examples. Training gets very slow, and other methods do the same job faster.

Key takeaways

  1. 1
    Widest gap. Keep far from both groups.
  2. 2
    Edge points decide. The support vectors set the line.
  3. 3
    Curves too. A trick allows bent borders.

Learn it properly