Thursday 6 April 2017

Logistic Regression

What is Logistic Regression?

What can be types of predictors (independent variables) and dependent variable?
Predictors can be continuous, categorical or mix of both.
Dependent variable is categorical (nominal), binary or multinomial.

What is Binomial LR? Multinomial LR? Give some real world examples.

When to use Logistic Regression?

Write hypothesis.

What is logistic function?

What does logistic function actually do?

Write cost function.

Explain how cost function adds cost (error) in case when hypothesis predicts wrong outcome and how it handles the case when hypothesis predicts right value.

Write regularized cost function.

References:
https://www.r-bloggers.com/how-to-perform-a-logistic-regression-in-r/
https://github.com/BojanKomazec/machine-learning-stanford-course/blob/master/machine-learning-ex2/ex2.pdf



No comments:

Post a Comment