Hausdorff clustering

TL;DR

This paper introduces a clustering algorithm based on the Hausdorff distance, demonstrating its effectiveness in distinguishing complex structures compared to traditional linkage methods through applications to toy and financial data.

Contribution

The paper presents a mathematically grounded Hausdorff clustering method and compares its performance to single and complete linkage techniques.

Findings

Hausdorff clustering effectively discriminates complex structures.

It performs well on toy and financial time series data.

Hausdorff linkage offers a robust alternative to traditional methods.

Abstract

A clustering algorithm based on the Hausdorff distance is introduced and compared to the single and complete linkage. The three clustering procedures are applied to a toy example and to the time series of financial data. The dendrograms are scrutinized and their features confronted. The Hausdorff linkage relies of firm mathematical grounds and turns out to be very effective when one has to discriminate among complex structures.