High AUC does not guarantee a defect prediction model outperforms random guessing on true and false positive rates for all thresholds.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SE 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Evaluating Software Defect Prediction Models via the Area Under the ROC Curve Can Be Misleading
High AUC does not guarantee a defect prediction model outperforms random guessing on true and false positive rates for all thresholds.