Abstract: One of the most commonly used algorithms for classification of health datasets is the K-Nearest Neighbor (KNN). However, KNN performs poorly when dealing with data that has many features.