This is a test version of Biostars. For the public version, visit https://www.biostars.org.
how to compare the cluster quality of k means algorithm and k harmonic means algorithm.

I want to compare both k means and k harmonic means algorithm in terms of cluster quality. Can anyone suggest a good measure for comparison. Thanks in advance.

cluster-validation k-means kharmonic-means
• 3,204 views • link
updated by Ram Ram • written by sithara.ep • sithara.ep

1 answer

Most cluster metrics will be some form of inter. vs intra. cluster distances. The gist of these metrics is to see how how close are members within a cluster and how different are members among clusters. There really isn't a single good metric.

It depends on the input data and what you are looking for in clustering. The best way to compare cluster algorithms is to use real data for benchmarking.

• 0 views • link
updated by Ram Ram • written by Damian Kao Damian Kao

In my project I am trying to combine both k harmonic mean and Artificial bee colony algorithm.k harmonic mean is used for calculating the fitness value of Artificial bee colony algorithm. And I hope it will improve the quality of clustering. Can you say I am on the right track? please help me if you are working in this area.

• 0 views • link
updated by Ram Ram • written by sithara.ep • sithara.ep

Log in to answer this question.

More posts like this