A KNN model performs well on a dataset with 5 features but poorly after adding 200 irrelevant random features. Which phenomenon explains this?

Question

Answer

Underfitting — the model needs more neighbours to handle more features

Answer

Curse of dimensionality — in high dimensions, distances become nearly uniform, making nearest neighbours meaningless

Answer

Feature leakage — the random features contain target information

Answer

Overfitting — KNN memorised the 200 new features

Crowdly