https://answer.freetools.me/tags/roc/ Recent content in ROC on Answer Hugo -- 0.152.2 zh-cn Thu, 12 Mar 2026 21:08:27 +0800 https://answer.freetools.me/%E5%BD%9399%E5%87%86%E7%A1%AE%E7%8E%87%E6%88%90%E4%B8%BA%E8%B0%8E%E8%A8%80%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0%E8%AF%84%E4%BC%B0%E6%8C%87%E6%A0%87%E7%9A%84%E6%B7%B1%E5%B1%82%E5%8D%9A%E5%BC%88/ Thu, 12 Mar 2026 21:08:27 +0800 https://answer.freetools.me/%E5%BD%9399%E5%87%86%E7%A1%AE%E7%8E%87%E6%88%90%E4%B8%BA%E8%B0%8E%E8%A8%80%E6%9C%BA%E5%99%A8%E5%AD%A6%E4%B9%A0%E8%AF%84%E4%BC%B0%E6%8C%87%E6%A0%87%E7%9A%84%E6%B7%B1%E5%B1%82%E5%8D%9A%E5%BC%88/ 从混淆矩阵的数学根基出发，系统解析准确率、精确率、召回率、F1-score等分类指标的本质与局限，深入剖析ROC/PR曲线与AUC的工作原理，揭示为何准确率会在不平衡数据上"说谎"，以及医疗诊断、欺诈检测等场景中如何正确选择评估指标。