Nearest Neighbor Classifiers: Reducing the Computational Demands
R. Raj Kumar, P. Viswanath, Chigarapalle Shoba Bindu · 2016
Nearest Neighbor Classifiers demand high computational resources i.e, time and memory. Two distinct methods are followed by researchers in Pattern Recognition to reduce this computational burden. The first method is reducing the reference set or training set and the second method is dimensionality reduction which are referred as Prototype Selection and Feature Reduction(a.k.a Feature Extraction or Feature selection) respectively. In this paper, we cascaded the two methods to reduce the data set in both directions there by reducing the computational burden of Nearest Neighbor Classifier. The experiments are done on the bench mark datasets and the results obtained are satisfactory.