Classification using distance nearest neighbours

TL;DR

This paper introduces a probabilistic classification method based on Markov random fields that models class label dependencies using feature vector distances, offering an efficient alternative to existing approaches.

Contribution

It presents a novel, computationally efficient algorithm for probabilistic classification using Markov random fields and distance-based dependencies, improving upon prior methods.

Findings

Encouraging results compared to k-nearest neighbor algorithm

More efficient computational algorithm for Markov random field model

Probabilistic basis for statistical inference in classification

Abstract

This paper proposes a new probabilistic classification algorithm using a Markov random field approach. The joint distribution of class labels is explicitly modelled using the distances between feature vectors. Intuitively, a class label should depend more on class labels which are closer in the feature space, than those which are further away. Our approach builds on previous work by Holmes and Adams (2002, 2003) and Cucala et al. (2008). Our work shares many of the advantages of these approaches in providing a probabilistic basis for the statistical inference. In comparison to previous work, we present a more efficient computational algorithm to overcome the intractability of the Markov random field model. The results of our algorithm are encouraging in comparison to the k-nearest neighbour algorithm.