8. Data Leakage

Data leakage is a serious bane in machine learning, which usually results in overly optimistic model results.

8.1. Examples

Some subtle examples of data leakages.

_images/dataleakage.png

University of Michigan: Coursera Data Science in Python

8.2. Types of Leakages

Data Leakages can be classified into two.

_images/dataleakage2.png

University of Michigan: Coursera Data Science in Python

8.3. Detecting Leakages

_images/dataleakage3.png

University of Michigan: Coursera Data Science in Python

8.4. Minimising Leakages

_images/dataleakage4.png

University of Michigan: Coursera Data Science in Python