The Altair Community is migrating to a new platform to provide a better experience for you. In preparation for the migration, the Altair Community is on read-only mode from October 28 - November 6, 2024. Technical support via cases will continue to work as is. For any urgent requests from Students/Faculty members, please submit the form linked here
"Numeric Dimension Reduction (PCA, etc) for Cluster Analysis"
Contributor II
I'm working on a clustering application and want to reduce the numeric attributes, especially correlated ones, prior to clustering. A proven technique to doing so is PCA, which RapidMiner contains. What I also want to do is the Varimax rotation algorithm to "load" a minimum number of attributes per component to improve interpretability. In a prior post, it was mentioned that PCA rotation (Varimax, etc.) is on the future wish list. I have the following questions on this:
1.) How close is RapidMiner to including rotation in future releases?
2.) If not "close", would I be better off looking at using an R package within RapidMiner to accomplish this?
3.) What other techniques have people used in RM to reduce numeric dimensions to a minimum, non-correlated set that are just as good/better than PCA for cluster analysis?
Thanks!
1.) How close is RapidMiner to including rotation in future releases?
2.) If not "close", would I be better off looking at using an R package within RapidMiner to accomplish this?
3.) What other techniques have people used in RM to reduce numeric dimensions to a minimum, non-correlated set that are just as good/better than PCA for cluster analysis?
Thanks!
1
