News for hackers. Stuff that matters.
Re:Machine Learning Cheat Sheet for scikit-learn by tlarkworthy on Thursday June 06, 2013 @08:37AM (#5831982)
" find the number of clusters"

Plot k against prediction accuracy (averaged over a number of runs). Look for a step, that is the correct setting of k. I don't believe there exists an algorithm for detecting that step as good as I can.

Re:Machine Learning Cheat Sheet for scikit-learn by ogrisel on Thursday June 06, 2013 @10:55AM (#5832765)
Prediction accuracy cannot be computed in a purely unsupervised setting: you don't have labels for the samples.

You can compare the cluster found by k-means run several times against what you get with a randomized version of your dataset though: http://blog.echen.me/2011/03/19/counting-clusters/

Re:Machine Learning Cheat Sheet for scikit-learn by dk8996 on Thursday June 06, 2013 @10:37AM (#5832669)
I think its called x-means algo. Here is the paper http://www.cs.cmu.edu/~dpelleg/download/xmeans.pdf