Final Thoughts

Advantages

Disadvantages

It's simple to understand and explain.

All training data must be stored.

Model training is fast.

• KNN is a lazy learner, and the algorithm does not learn from the training data. It memorizes the training dataset.

The prediction phase can be slow when test data is large.

It can be used for classification and regression.

It's sensitive to irrelevant features and feature scaling.

It’s a nonlinear algorithm:

• The decision boundaries of KNN are locally linear segments but generally have a complex shape that is not equivalent to a line in 2D or a hyperplane in higher dimensions.

Accuracy is (generally) not competitive with the best supervised learning methods.