An Appraise of KNN to the Perfection

Abstract

K-Nearest Neighbor (KNN) is highly efficient classification algorithm due to its key features like: very easy to use, requires low training time, robust to noisy training data, easy to implement. However, it also has some shortcomings like high computational complexity, large memory requirement for large training datasets, curse of dimensionality and equal weights given to all attributes. Many researchers have suggested various advancements and improvements in KNN to overcome these shortcomings. This paper appraising various advancements and improvements in KNN.