A Data Scientist uses logistic regression to build a fraud detection model. While the model accuracy is 99%, 90% of the fraud cases are not detected by the model. What action will definitively help…

The correct answer is B. Decreasing the class probability threshold. Decreasing the class probability threshold makes the model more sensitive and, therefore, marks more cases as the positive class, which is fraud in this case. This will increase the likelihood of detection. However, it comes at the price of lowering precision.

Modeling

Question

A Data Scientist uses logistic regression to build a fraud detection model. While the model accuracy is 99%, 90% of the fraud cases are not detected by the model. What action will definitively help the model detect more than 10% of fraud cases?

Options

AUsing undersampling to balance the dataset
BDecreasing the class probability threshold
CUsing regularization to reduce overfitting
DUsing oversampling to balance the dataset

How the community answered

(42 responses)

A
2% (1)
B
79% (33)
C
14% (6)
D
5% (2)

Explanation

Decreasing the class probability threshold makes the model more sensitive and, therefore, marks more cases as the positive class, which is fraud in this case. This will increase the likelihood of detection. However, it comes at the price of lowering precision.

Topics

#Classification Threshold#Recall#Logistic Regression#Imbalanced Data

Community Discussion

No community discussion yet for this question.

Full MLS-C01 Practice