Specificity, sensitivity and AUC measures via RM v9.1

CindyDK
CindyDK New Altair Community Member
edited November 5 in Community Q&A
For non-binominal/polynominal label target data, what would be the best/recommended cross-validation testing data performance measures/operators for Specificity, sensitivity, and AUC measures in RM v9.1?
Tagged:

Best Answers

  • Telcontar120
    Telcontar120 New Altair Community Member
    Answer ✓
    ROC / AUC and its related measures are designed for binominal label classification problems only.
    The best way to approach if you have multiple classes is to encode with a one-vs-all-others approach.  So if you have possible label values {1,2,3} then encode that as 1 vs not 1, and generate the ROC.  You can repeat for the other classes as needed, store the performances, and average them as well if you want.

Answers

  • Telcontar120
    Telcontar120 New Altair Community Member
    Answer ✓
    ROC / AUC and its related measures are designed for binominal label classification problems only.
    The best way to approach if you have multiple classes is to encode with a one-vs-all-others approach.  So if you have possible label values {1,2,3} then encode that as 1 vs not 1, and generate the ROC.  You can repeat for the other classes as needed, store the performances, and average them as well if you want.
  • CindyDK
    CindyDK New Altair Community Member
    Thanks for the sharing. It seems to utilize the loop to cover the number of the classes (Re: 1 vs all others). Any other thoughts about "specificity, sensitivity" measures?