A KNN classifier with K=1 achieves 98% training accuracy but only 72% test accuracy. A KNN classifier with K=25 achieves 80% training accuracy and 79% test accuracy. What is the most accurate diagnosis?

Question

Answer

K=1 is the better model because training accuracy is more reliable than test accuracy

Answer

K=25 is underfitting because training accuracy dropped below 90%

Answer

Both models have the same bias-variance profile since both are KNN

Answer

K=1 is overfitting; K=25 has better generalisation but may still be suboptimal

Crowdly

A KNN classifier with K=1 achieves 98% training accuracy but only 72% test accur...

Want instant access to all verified answers on moodle.ucu.ac.ug?