Unverified50% confidenceFactExact time
2006年Janez Demšar发表论文《Statistical Comparisons of Classifiers over Multiple Data Sets》，论证配对t检验在比较分类器时存在问题，并推荐使用Wilcoxon符号秩检验或Friedman检验配合Nemenyi事后检验
1
Sources
50%
Confidence
Long-term
Relevance
8/9/2026
First Seen
Sources
Related Claims
Unverified多种子重复实验中可使用配对t检验或Wilcoxon符号秩检验比较两组准确率分布，其中Wilcoxon检验因不依赖正态性假设更适合小样本74% similarUnverified当比较两个具体训练模型的预测行为差异时应固定种子使用同一测试集运行McNemar检验，当评估方法整体优越性时应用多种子重复实验配合配对t检验或非参数检验61% similarUnverified该项目采用配对McNemar检验作为统计显著性验证方法61% similarUnverifiedMutation Testing checks whether test cases are truly effective by deliberately introducing defects into the code.61% similarUnverifiedThe author benchmarked 8 classifiers, including logistic regression, random forest, XGBoost, and LightGBM.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/719116API
curl https://kongchang.com/api/v1/knowledge/claims/719116MCP
get_claim(id=719116)