k-Nearest Neighbors

Type
Classification

The Elbow Curve for parameter selection:

Properties of a distance function:

1. $d(x,y)>0$ (when $x\ne y$)
2. $d(x,y)=d(y,x)$
3. $d(x,z)\le d(x,y)+d(y,z)$ (triangle inequality)

Distance Function Examples:

• Euclidian Distance
• Minkowski Distance
• Manhattan Distance