Question 1

Why not use Linear Regression for classification?

Accepted Answer

Linear regression can output values greater than 1 or less than 0, making it unsuitable for probabilities. It is also sensitive to outliers.

Question 2

What is the Sigmoid curve shape?

Accepted Answer

It is an S-shaped curve that asymptoticly approaches 0 and 1.

Question 3

How do you evaluate a classification model?

Accepted Answer

Using a Confusion Matrix, Accuracy, Precision, Recall, and AUC-ROC score.

Question 4

What is a false positive?

Accepted Answer

When the model incorrectly predicts class 1 (positive) when the actual value is class 0 (negative).

Why not use Linear Regression for classification?

More FAQs in What is Logistic Regression? Binary Classification Explained