How to decide the quality of a distance measure in machine learning techniques?

Question

How can I decide if a given distance metric is a good?

Also, please give us some context. What exactly is the problem you're trying to tackle using machine learning? — NPE, Mar 21 '13 at 10:03

score 1 · Answer 1 · answered Mar 25 '13 at 19:11

Classification or clustering?

For classification, use whichever gives the highest cross-validated accuracy (or whichever quality measure you need).

For clustering, you need to pick a criteria for cluster quality, and choose the one that gives you the best cluster quality:

score 0 · Answer 2 · answered Mar 21 '13 at 13:24

0

The metric is good, if it works for your problem.

In classification, it's as simple as that. If it gives good results, go with it. Otherwise, try a different metric.

answered Mar 21 '13 at 13:24

2 Answers2