K-Means Kernel Classifier

TL;DR

This paper introduces a novel classification method that combines K-means clustering with least-squares kernel classification, using representative vectors for improved accuracy demonstrated on MNIST.

Contribution

The paper presents a new hybrid approach that integrates unsupervised clustering with supervised kernel classification, enhancing classification performance.

Findings

High accuracy on MNIST dataset

Effective use of representative vectors for classification

Combines unsupervised and supervised learning successfully

Abstract

We combine K-means clustering with the least-squares kernel classification method. K-means clustering is used to extract a set of representative vectors for each class. The least-squares kernel method uses these representative vectors as a training set for the classification task. We show that this combination of unsupervised and supervised learning algorithms performs very well, and we illustrate this approach using the MNIST dataset