BookmarkSubscribeRSS Feed

Using the ROC Convex Hull Algorithm to Select a Classifier

Started ‎12-06-2021 by
Modified ‎12-07-2021 by
Views 4,479

Rick Wicklin (@Rick_SAS)  contributed an article in The Do Loop (https://blogs.sas.com/content/iml/2021/11/15/2d-convex-hull-sas.html) describing two-dimensional complex hulls in SAS, specifically centered on the SAS/IML CVXHULL function.

 

A while ago, I created an algorithm based on Provost and Foster's work using a convex hull representation of several classifiers to select the optimal classifier based on cost-weighted classification accuracy. I presented my work at the Interface 2003 conference under the title "Cost-Sensitive Classifier Selection Using the ROC Convex Hull Method" and wish to share it with the SAS community. The paper and the SAS macro are included (attached). I hope that they will be of interest to some SAS community members.

Version history
Last update:
‎12-07-2021 07:27 AM
Updated by:

sas-innovate-white.png

Our biggest data and AI event of the year.

Don’t miss the livestream kicking off May 7. It’s free. It’s easy. And it’s the best seat in the house.

Join us virtually with our complimentary SAS Innovate Digital Pass. Watch live or on-demand in multiple languages, with translations available to help you get the most out of every session.

 

Register now!

SAS AI and Machine Learning Courses

The rapid growth of AI technologies is driving an AI skills gap and demand for AI talent. Ready to grow your AI literacy? SAS offers free ways to get started for beginners, business leaders, and analytics professionals of all skill levels. Your future self will thank you.

Get started

Article Tags