8. You've just finished fitting a logistic regression model for email spam detection, and it is getting AUC of 0.95 on Train and 0.60 on Validation dataset. You know that your implementation has no bugs, so what could be causing the problem? (2 points) a. You are overfitting b. You are underfitting 9. Choose True or False? (2 points) a. (True/False) Size of the 95% confidence interval for the sample mean decreases as the as the sample size increases, if all else remains the same, b. (True/False) Multicollinearity is a big problem if individual regression coefficients are of interest and the goal of the analysis is to reliably explanation the relationship of a predictor with the outcome