Particle swarm optimisation-based KNN for improving KNN and ensemble classification performance

摘要: In this paper we use KNN algorithm for our ensemble classification process that finds out the nearest K training samples given a test sample where each is predictions vector generated by combined computation of classifiers and algebraic combiners. attempt to reduce computational time involved in finding KNN, used particle swarm optimisation PSO with which randomly selects from set until global consensus reached among particles label an appropriate class weighted majority voting WMV samples. The proposed method demonstrated better performance as compared other traditional methods terms generalisation accuracies when tested over several datasets UCI repository high dimensional datasets.