Support Vector Machine
In plain terms
A classifier that finds the optimal boundary between classes.
Going deeper
An SVM finds the hyperplane that best separates classes with the widest margin, using kernels to handle non-linear data. It maximizes the gap between groups.
Under the hood
The kernel trick maps data to higher dimensions; effective in high dimensions but scales poorly to huge datasets.
Examples
Text classification, image recognition, small datasets, kernel methods.
Think of it like...
Like drawing the widest possible no-man's-land street between two neighborhoods.