近日使用R语言对某个数据集进行logit回归建模,所用数据是经过清洗后的,但建型建立后出现说几个特征异常,没有特征系数前来前教各位前辈，到底是算法包本身的问题，还是我的数据本身的问题？(已上传数据集至附件）
lgfit<-glm(Churn~.,family = binomial(link ='logit'),data=traindata)
summary(lgfit) # 四个变量无系数,说它们异常,实际上数据并没问题,估计是算法包本身的问题

Call:
glm(formula = Churn ~ ., family = binomial(link = "logit"), data = traindata)

Coefficients: (4 not defined because of singularities)
Estimate Std. Error z value Pr(>|z|)
(Intercept) 3.529e+075.037e+07 0.701 0.483505
Gender女 1.338e-018.495e-02 1.575 0.115153
HandsetASAD90 5.560e+002.788e-0119.942< 2e-16 ***
Usage_Band中使用率 -8.563e-012.082e-01-4.113 3.90e-05 ***
International_mins 1.496e-023.884e-03 3.850 0.000118 ***
...
National_calls NA NA NA NA
National_mins NA NA NA NA
All_calls_mins NA NA NA NA
Nat_call_cost 1.133e-022.560e-02 0.443 0.657953
Weekend_mins_Fluctuation1.050e-032.406e-02 0.044 0.965196
...
Mins_charge NA NA NA NA
actual_call_cost 2.481e-022.773e-02 0.895 0.371031
Total_call_cost 3.366e+054.797e+05 0.702 0.482842
Total_Cost -3.366e+054.797e+05-0.702 0.482842
call_cost_per_min -3.940e-026.678e-02-0.590 0.555170
average_cost_min -1.450e+005.689e-01-2.548 0.010829 *
---
Signif. codes:0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1