A pseudo nearest centroid neighbour classifier

Abstract

In this paper, we propose a new reliable classification approach, called the pseudo nearest centroid neighbour rule, which is based on the pseudo nearest neighbour rule (PNN) and nearest centroid neighbourhood (NCN). In the proposed PNCN, the nearest centroid neighbours rather than nearest neighbours per class are first searched by means of NCN. Then, we calculate k categorical local mean vectors corresponding to k nearest centroid neighbours, and assign a weight to each local mean vector. Using the weighted k local mean vectors for each class, PNCN designs the corresponding pseudo nearest centroid neighbour and decides the class label of the query pattern according to the closest pseudo nearest centroid neighbour among all classes. The classification performance of the proposed PNCN is evaluated on real and artificial datasets in terms of the classification accuracy. The experimental results demonstrate the effectiveness and robustness of PNCN over the competing methods in many practical classification problems.