为什么对于多个模型,我的精确度很高,但ROC AUC很低?
我的数据集大小是42542 x 14,我正在尝试建立不同的模型,如逻辑回归、KNN、RF、决策树,并比较精确度 我得到了每个模型的高精度但低ROC AUC 数据中约有85%的样本的目标变量为1,15%的样本的目标变量为0。为了处理这种不平衡,我试着取样,但结果还是一样的 glm的系数如下:为什么对于多个模型,我的精确度很高,但ROC AUC很低?,r,model,logistic-regression,auc,R,Model,Logistic Regression,Auc,我的数据集大小是42542 x 14,我正在尝试建立不同的模型,如逻辑回归、KNN、RF、决策树,并比较精确度 我得到了每个模型的高精度但低ROC AUC 数据中约有85%的样本的目标变量为1,15%的样本的目标变量为0。为了处理这种不平衡,我试着取样,但结果还是一样的 glm的系数如下: glm(formula = loan_status ~ ., family = "binomial", data = lc_train) Deviance Residuals: Min
glm(formula = loan_status ~ ., family = "binomial", data = lc_train)
Deviance Residuals:
Min 1Q Median 3Q Max
-2.7617 0.3131 0.4664 0.6129 1.6734
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -8.264e+00 8.338e-01 -9.911 < 2e-16 ***
annual_inc 5.518e-01 3.748e-02 14.721 < 2e-16 ***
home_own 4.938e-02 3.740e-02 1.320 0.186780
inq_last_6mths1 -2.094e-01 4.241e-02 -4.938 7.88e-07 ***
inq_last_6mths2-5 -3.805e-01 4.187e-02 -9.087 < 2e-16 ***
inq_last_6mths6-10 -9.993e-01 1.065e-01 -9.380 < 2e-16 ***
inq_last_6mths11-15 -1.448e+00 3.510e-01 -4.126 3.68e-05 ***
inq_last_6mths16-20 -2.323e+00 7.946e-01 -2.924 0.003457 **
inq_last_6mths21-25 -1.399e+01 1.970e+02 -0.071 0.943394
inq_last_6mths26-30 1.039e+01 1.384e+02 0.075 0.940161
inq_last_6mths31-35 -1.973e+00 1.230e+00 -1.604 0.108767
loan_amnt -1.838e-05 3.242e-06 -5.669 1.43e-08 ***
purposecredit_card 3.286e-02 1.130e-01 0.291 0.771169
purposedebt_consolidation -1.406e-01 1.032e-01 -1.362 0.173108
purposeeducational -3.591e-01 1.819e-01 -1.974 0.048350 *
purposehome_improvement -2.106e-01 1.189e-01 -1.771 0.076577 .
purposehouse -3.327e-01 1.917e-01 -1.735 0.082718 .
purposemajor_purchase -7.310e-03 1.288e-01 -0.057 0.954732
purposemedical -4.955e-01 1.530e-01 -3.238 0.001203 **
purposemoving -4.352e-01 1.636e-01 -2.661 0.007800 **
purposeother -3.858e-01 1.105e-01 -3.493 0.000478 ***
purposerenewable_energy -8.150e-01 3.036e-01 -2.685 0.007263 **
purposesmall_business -9.715e-01 1.186e-01 -8.191 2.60e-16 ***
purposevacation -4.169e-01 2.012e-01 -2.072 0.038294 *
purposewedding 3.909e-02 1.557e-01 0.251 0.801751
open_acc -1.408e-04 4.147e-03 -0.034 0.972923
gradeB -4.377e-01 6.991e-02 -6.261 3.83e-10 ***
gradeC -5.858e-01 8.340e-02 -7.024 2.15e-12 ***
gradeD -7.636e-01 9.558e-02 -7.990 1.35e-15 ***
gradeE -7.832e-01 1.115e-01 -7.026 2.13e-12 ***
gradeF -9.730e-01 1.325e-01 -7.341 2.11e-13 ***
gradeG -1.031e+00 1.632e-01 -6.318 2.65e-10 ***
verification_statusSource Verified 6.340e-02 4.435e-02 1.429 0.152898
verification_statusVerified 6.864e-02 4.400e-02 1.560 0.118739
dti -4.683e-03 2.791e-03 -1.678 0.093373 .
fico_range_low 6.705e-03 9.292e-04 7.216 5.34e-13 ***
term 5.773e-01 4.499e-02 12.833 < 2e-16 ***
emp_length2-4 years 6.341e-02 4.911e-02 1.291 0.196664
emp_length5-9 years -3.136e-02 5.135e-02 -0.611 0.541355
emp_length10+ years -2.538e-01 5.185e-02 -4.895 9.82e-07 ***
delinq_2yrs2+ 5.919e-02 9.701e-02 0.610 0.541754
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 25339 on 29779 degrees of freedom
Residual deviance: 23265 on 29739 degrees of freedom
AIC: 23347
Number of Fisher Scoring iterations: 10
Confusion Matrix and Statistics
Reference
Prediction 0 1
0 32 40
1 1902 10788
Accuracy : 0.8478
95% CI : (0.8415, 0.854)
No Information Rate : 0.8485
P-Value [Acc > NIR] : 0.5842
Kappa : 0.0213
Mcnemar's Test P-Value : <2e-16
Sensitivity : 0.016546
Specificity : 0.996306
Pos Pred Value : 0.444444
Neg Pred Value : 0.850118
Prevalence : 0.151544
Detection Rate : 0.002507
Detection Prevalence : 0.005642
Balanced Accuracy : 0.506426
'Positive' Class : 0
glm(公式=贷款状态~,family=“二项式”,数据=信用证列车)
偏差残差:
最小1季度中值3季度最大值
-2.7617 0.3131 0.4664 0.6129 1.6734
系数:
估计标准误差z值Pr(>z)
(截距)-8.264e+00 8.338e-01-9.911<2e-16***
年鉴公司5.518e-01 3.748e-02 14.721<2e-16***
home_own 4.938e-02 3.740e-02 1.320 0.186780
inq_last_6mths1-2.094e-01 4.241e-02-4.938 7.88e-07***
inq_last_6mths2-5-3.805e-01 4.187e-02-9.087<2e-16***
inq_last_6mths6-10-9.993e-01 1.065e-01-9.380<2e-16***
inq_last_6mths11-15-1.448e+00 3.510e-01-4.126 3.68e-05***
inq_last_6mths16-20-2.323e+00 7.946e-01-2.924 0.003457**
inq_last_6mths21-25-1.399e+01 1.970e+02-0.071 0.943394
inq_last_6mths26-30 1.039e+01 1.384e+02 0.075 0.940161
最后6个月31-35-1.973e+001.230e+00-1.6040.108767
贷款金额-1.838e-05 3.242e-06-5.669 1.43e-08***
目的信用卡3.286e-02 1.130e-01 0.291 0.771169
目的债务合并-1.406e-01 1.032e-01-1.362 0.173108
教育目的-3.591e-01 1.819e-01-1.974 0.048350*
目的家庭改善-2.106e-01 1.189e-01-1.771 0.076577。
目的地-3.327e-01 1.917e-01-1.7350.082718。
目的主要采购-7.310e-03 1.288e-01-0.057 0.954732
目的医学-4.955e-01 1.530e-01-3.238 0.001203**
目的-4.352e-01 1.636e-01-2.661 0.007800**
目的其他-3.858e-01 1.105e-01-3.493 0.000478***
目的可再生能源-8.150e-01 3.036e-01-2.685 0.007263**
目的小型企业-9.715e-01 1.186e-01-8.191 2.60e-16***
目的-4.169e-01 2.012e-01-2.072 0.038294*
目的婚礼3.909e-02 1.557e-01 0.251 0.801751
打开附件-1.408e-04 4.147e-03-0.034 0.972923
B级-4.377e-01 6.991e-02-6.261 3.83e-10***
C级-5.858e-01 8.340e-02-7.024 2.15e-12***
等级-7.636e-01 9.558e-02-7.990 1.35e-15***
等级E-7.832e-01 1.115e-01-7.026 2.13e-12***
F级-9.730e-01 1.325e-01-7.341 2.11e-13***
G级-1.031e+00 1.632e-01-6.318 2.65e-10***
验证\u状态来源验证6.340e-02 4.435e-02 1.429 0.152898
验证状态验证6.864e-02 4.400e-02 1.560 0.118739
dti-4.683e-03 2.791e-03-1.678 0.093373。
fico_范围_低6.705e-03 9.292e-04 7.216 5.34e-13***
术语5.773e-01 4.499e-02 12.833<2e-16***
emp_长度2-4年6.341e-02 4.911e-02 1.291 0.196664
emp_长度5-9年-3.136e-02 5.135e-02-0.611 0.541355
emp_长度10+年-2.538e-01 5.185e-02-4.895 9.82e-07***
delinq_2yrs2+5.919e-02 9.701e-02 0.610 0.541754
---
签名。代码:0'***'0.001'***'0.01'*'0.05'.'0.1''1
(二项式族的离散参数取为1)
零偏差:29779自由度上的25339
剩余偏差:29739自由度上的23265
工商行政管理局:23347
Fisher评分迭代次数:10
LR的混淆矩阵如下所示:
glm(formula = loan_status ~ ., family = "binomial", data = lc_train)
Deviance Residuals:
Min 1Q Median 3Q Max
-2.7617 0.3131 0.4664 0.6129 1.6734
Coefficients:
Estimate Std. Error z value Pr(>|z|)
(Intercept) -8.264e+00 8.338e-01 -9.911 < 2e-16 ***
annual_inc 5.518e-01 3.748e-02 14.721 < 2e-16 ***
home_own 4.938e-02 3.740e-02 1.320 0.186780
inq_last_6mths1 -2.094e-01 4.241e-02 -4.938 7.88e-07 ***
inq_last_6mths2-5 -3.805e-01 4.187e-02 -9.087 < 2e-16 ***
inq_last_6mths6-10 -9.993e-01 1.065e-01 -9.380 < 2e-16 ***
inq_last_6mths11-15 -1.448e+00 3.510e-01 -4.126 3.68e-05 ***
inq_last_6mths16-20 -2.323e+00 7.946e-01 -2.924 0.003457 **
inq_last_6mths21-25 -1.399e+01 1.970e+02 -0.071 0.943394
inq_last_6mths26-30 1.039e+01 1.384e+02 0.075 0.940161
inq_last_6mths31-35 -1.973e+00 1.230e+00 -1.604 0.108767
loan_amnt -1.838e-05 3.242e-06 -5.669 1.43e-08 ***
purposecredit_card 3.286e-02 1.130e-01 0.291 0.771169
purposedebt_consolidation -1.406e-01 1.032e-01 -1.362 0.173108
purposeeducational -3.591e-01 1.819e-01 -1.974 0.048350 *
purposehome_improvement -2.106e-01 1.189e-01 -1.771 0.076577 .
purposehouse -3.327e-01 1.917e-01 -1.735 0.082718 .
purposemajor_purchase -7.310e-03 1.288e-01 -0.057 0.954732
purposemedical -4.955e-01 1.530e-01 -3.238 0.001203 **
purposemoving -4.352e-01 1.636e-01 -2.661 0.007800 **
purposeother -3.858e-01 1.105e-01 -3.493 0.000478 ***
purposerenewable_energy -8.150e-01 3.036e-01 -2.685 0.007263 **
purposesmall_business -9.715e-01 1.186e-01 -8.191 2.60e-16 ***
purposevacation -4.169e-01 2.012e-01 -2.072 0.038294 *
purposewedding 3.909e-02 1.557e-01 0.251 0.801751
open_acc -1.408e-04 4.147e-03 -0.034 0.972923
gradeB -4.377e-01 6.991e-02 -6.261 3.83e-10 ***
gradeC -5.858e-01 8.340e-02 -7.024 2.15e-12 ***
gradeD -7.636e-01 9.558e-02 -7.990 1.35e-15 ***
gradeE -7.832e-01 1.115e-01 -7.026 2.13e-12 ***
gradeF -9.730e-01 1.325e-01 -7.341 2.11e-13 ***
gradeG -1.031e+00 1.632e-01 -6.318 2.65e-10 ***
verification_statusSource Verified 6.340e-02 4.435e-02 1.429 0.152898
verification_statusVerified 6.864e-02 4.400e-02 1.560 0.118739
dti -4.683e-03 2.791e-03 -1.678 0.093373 .
fico_range_low 6.705e-03 9.292e-04 7.216 5.34e-13 ***
term 5.773e-01 4.499e-02 12.833 < 2e-16 ***
emp_length2-4 years 6.341e-02 4.911e-02 1.291 0.196664
emp_length5-9 years -3.136e-02 5.135e-02 -0.611 0.541355
emp_length10+ years -2.538e-01 5.185e-02 -4.895 9.82e-07 ***
delinq_2yrs2+ 5.919e-02 9.701e-02 0.610 0.541754
---
Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1
(Dispersion parameter for binomial family taken to be 1)
Null deviance: 25339 on 29779 degrees of freedom
Residual deviance: 23265 on 29739 degrees of freedom
AIC: 23347
Number of Fisher Scoring iterations: 10
Confusion Matrix and Statistics
Reference
Prediction 0 1
0 32 40
1 1902 10788
Accuracy : 0.8478
95% CI : (0.8415, 0.854)
No Information Rate : 0.8485
P-Value [Acc > NIR] : 0.5842
Kappa : 0.0213
Mcnemar's Test P-Value : <2e-16
Sensitivity : 0.016546
Specificity : 0.996306
Pos Pred Value : 0.444444
Neg Pred Value : 0.850118
Prevalence : 0.151544
Detection Rate : 0.002507
Detection Prevalence : 0.005642
Balanced Accuracy : 0.506426
'Positive' Class : 0
混淆矩阵与统计
参考文献
预测0 1
0 32 40
1 1902 10788
准确度:0.8478
95%可信区间:(0.8415,0.854)
无信息率:0.8485
P值[Acc>NIR]:0.5842
卡帕值:0.0213
Mcnemar的测试P值:如果有人提出混淆矩阵并谈论低ROC AUC,通常意味着他/她已将预测/概率转换为0和1,而ROC AUC公式并不要求这样做-它对原始概率有效,从而产生更好的结果。如果目标是获得最佳AUC值,最好在培训时将其设置为评估指标,这样可以获得比其他指标更好的结果。
这似乎是关于数据建模的问题,而不是特定的编程问题。这些问题属于像or这样的站点,而不是堆栈溢出。嗨,sneha,我建议你画一些这样的图,这样你就可以理解很容易获得100%的准确度,困难的是保持平衡为什么你认为0.85是好/高准确度,如果不使用任何ML就可以获得相同的值,只是预测所有情况下的1?换句话说,loan_status~1
似乎和你复杂的模型一样好。它是不平衡的,你的模型试图预测一切都是1,因为在混乱矩阵中,(1902+10788)/(10788+40+32+1902)=0.99的预测是1,这比你在数据中看到的还要多;2.kappa值太低,表示