Majority of k nearest neighbors

Euclidean distance

Better classification ability

Entropy

(TP + TN) / (Total)

Mean Squared Error

Proportion of variance explained by the model

No convergence

Data shows a nonlinear relationship

Actual positive predicted as positive

Regression, Classification

Entropy, Information Gain, Gini Index

Value of k, Distance metric used, Size of the training set

Mean Absolute Error (MAE), Mean Squared Error (MSE), Root Mean Squared Error (RMSE)

True Positive, False Positive, False Negative, True Negative