Quantcast
Channel: Machine Learning
Viewing all articles
Browse latest Browse all 63107

Is there any way to rigorously validate a clustering method?

$
0
0

One can make visualizations of the clusters, plot the F statistic, etc, but they generally always improve as more clusters are included (increase k).

Should clustering always be just considered exploratory, and if so what would you do next?

submitted by lawrencechernin
[link][6 comments]

Viewing all articles
Browse latest Browse all 63107

Trending Articles