ROC and PR curve metrics derive from the composition G = F_p ∘ F_n^{-1} of class-conditional CDFs, yielding a geometric framework for understanding classifier behavior and operating point selection.
An introduction to ROC analysis
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
On the Geometry of Receiver Operating Characteristic and Precision-Recall Curves
ROC and PR curve metrics derive from the composition G = F_p ∘ F_n^{-1} of class-conditional CDFs, yielding a geometric framework for understanding classifier behavior and operating point selection.