Neural Network Clustering Based on Distances Between Objects

TL;DR

This paper introduces a distance-based clustering algorithm for high-dimensional data that uses neuron-like centers with lateral inhibition, producing consistent results independent of initial conditions.

Contribution

The paper proposes a novel clustering method that relies solely on object distances and employs neuron-inspired centers with lateral inhibition, ensuring stable and meaningful class detection.

Findings

Algorithm effectively finds class centers using only distance data.

Results are consistent regardless of initial conditions.

Computer simulations demonstrate the method's practical applicability.

Abstract

We present an algorithm of clustering of many-dimensional objects, where only the distances between objects are used. Centers of classes are found with the aid of neuron-like procedure with lateral inhibition. The result of clustering does not depend on starting conditions. Our algorithm makes it possible to give an idea about classes that really exist in the empirical data. The results of computer simulations are presented.