A dataset has a feature "Student ID" which is unique for every student. If Information Gain is used as the splitting criterion, what would happen?

Question

Answer

The algorithm would skip Student ID automatically because it detects it has too many unique values

Answer

Student ID would be treated identically to any other feature

Answer

Student ID would have zero Information Gain because it has no predictive power

Answer

Student ID would have the highest Information Gain because each split creates pure single-sample leaves

Crowdly