zekeLabs
Linear Models for Classification
“Goal - Become a Data Scientist”
“A Dream becomes a Goal when action is taken towards its achievement” - Bo Bennett
“The Plan”
“A Goal without a Plan is just a wish”
● Logistic Regression
● Need for Linear Discriminant Analysis
● Bias-Variance Tradeoff
● Correcting Bias-Variance Error
● Reducing Error using Learning algorithm
● Multicollinearity
● Impact of multicollinearity
● Detecting multicollinearity
● Correcting multicollinearity
Agenda
Logistic Regression - Limitations
● Logistic Regression is meant for binary classification & not often used for
multi-class classification
● Predicts unstable decision boundary when classes are well separated
● Unstable predictor with less training data
Understanding Variance
https://www.listendata.com/2014/11/detecting-and-correcting.html

Linear models of classification