Unverified80% confidenceFactTime unknown
ROC曲线在20世纪90年代才被机器学习研究者广泛用于分类模型的性能评估
1
Sources
80%
Confidence
Long-term
Relevance
6/1/2026
First Seen
Sources
Related Entities
Related Claims
Unverified假阳性与假阴性的权衡在信号检测理论中被形式化为ROC曲线（接收者操作特征曲线）60% similarUnverifiedWhen class imbalance is severe, the PR curve reflects the model's true performance better than the ROC curve.59% similarUnverifiedAUC（Area Under Curve）是衡量分类模型好坏最常用的指标之一58% similarUnverified2021年发表于《Nature Machine Intelligence》的一项系统综述分析了2010-2019年间COVID-19影像诊断相关的415个机器学习模型，发现大量模型因数据泄漏、不当的训练-测试划分或数据集偏差而无法在临床中实际使用58% similarUnverifiedAUC can be obtained by numerically integrating the area under the ROC curve using the trapezoidal rule, or equivalently computed via the Mann-Whitney U statistic.58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/20871API
curl https://kongchang.com/api/v1/knowledge/claims/20871MCP
get_claim(id=20871)