How would I identify important variables before doing any cluster analysis, I have a big data base of customer (purchase,loyalty..), 150 feature.
I want to work in two step: 1- Eliminate feature which is unuseful for clustering, e.g. an equivalent of correlation analyse between target and feature in supervised clustering. 2-Eliminate correlated feature, use cluster of variables, PCA. Then i do the classification.
I'm very well in the second step, but i have no idea for the first step, what is the equivalent in unsupervised clustering? which exploratory analysis can help me?
Check variable cluster analysis. proc varclus
You can also use the Variable Cluster Node under Explore
Catch the SAS Global Forum keynotes, announcements, and tech content!
sasglobalforum.com | #SASGF
Don't miss out on SAS Innovate - Register now for the FREE Livestream!
Can't make it to Vegas? No problem! Watch our general sessions LIVE or on-demand starting April 17th. Hear from SAS execs, best-selling author Adam Grant, Hot Ones host Sean Evans, top tech journalist Kara Swisher, AI expert Cassie Kozyrkov, and the mind-blowing dance crew iLuminate! Plus, get access to over 20 breakout sessions.
Use this tutorial as a handy guide to weigh the pros and cons of these commonly used machine learning algorithms.
Find more tutorials on the SAS Users YouTube channel.