machine learning key word (part two)

machine learning key word (part two)

error rate：错误率accuracy：精度error：误差training error：训练误差（或“empirical error：经验误差”）test error：测试误差generalization error：泛化误差overfitting：过拟合underfitting：欠拟合model selection：模型选择hold-out：留出法sampling：采样stratified sampling：分层采样cross validation：交叉验证k-fold cross validation：k折交叉验证validation set：验证集Leave-One-Out：留一法（简称“LOO”）bootstrapping：自助法bootstrap sampling：自助采样out-of-bag estimate：包外估计parameter tunning：调参performance measure：性能度量mean squared error：均方误差precision：查准率，亦称“准确率”recall：查全率，亦称“召回率”true positive：真正例True Positive Rate：真正例率（简称"TPR"）false positive：假正例False Positive Rate：假正例率（简称"FPR"）true negative：真反例false negative：假反例confusion matrix：混淆矩阵Break-Evendors Point：平衡点（简称“BEP”）threshold：阈值cut point：截断点ROC：受试者工作曲线AUC：Area under ROC Curveloss：损失unequal cost：非均等代价cost-sensitive：代价敏感hypothesis test：假设检验binomial test：二项测试confidence：置信度two-tailed：双边t-test：t检验paired t-test：成对t检验contingency table：列联表bias-variance decomposition：偏差-方差分解bias-variance dilemma：偏差-方差窘境cost-sensitive learning：代价敏感学习harmonic mean：调和平均