topic Re: Clustering Categorical Variables in Statistical Procedures

Clustering Categorical Variables

koomalkc — Fri, 13 Feb 2015 15:35:19 GMT

Hi,
Need some tips about how to do clustering for categorical variables.

Any epaper or thoughts are appreciated.

Thanks,

_kamal

Re: Clustering Categorical Variables

PGStats — Fri, 13 Feb 2015 16:07:01 GMT

You could use proc distance and choose a measure appropriate for nominal variables to get a distance matrix. Then use that distance matrix in proc cluster.

Re: Clustering Categorical Variables

SteveDenham — Fri, 13 Feb 2015 17:41:30 GMT

I wonder if you could get the output from PROC CORRESP, or the necessary internal datasets, and feed those to PROC CLUSTER.

Steve Denham

Re: Clustering Categorical Variables

PGStats — Fri, 13 Feb 2015 20:52:38 GMT

Proc Corresp calculates coordinates for the levels of two or more categorical variables based on their crosstabulation. If you fed distances derived from those coordinates to Proc Cluster, you could cluster together the levels of two or more categorical variables. I never tried doing that.

Re: Clustering Categorical Variables

koomalkc — Thu, 19 Feb 2015 15:04:53 GMT

Thanks guys

Re: Clustering Categorical Variables

turanbul — Tue, 18 Aug 2015 14:16:44 GMT

Hello,

I am also having a similar problem. As far as I can see using proc distance first and then apply Proc cluster seems appropriate.

What about in Miner? How can I do this with FASTCLUS and MODECLUS?