Separating Topological Noise from Features using Persistent Entropy

TL;DR

This paper introduces a simple method leveraging persistent entropy to distinguish topological noise from genuine features in data, enhancing topological data analysis accuracy.

Contribution

The paper proposes a novel measure called persistent entropy for comparing persistence barcodes, aiding in separating noise from features.

Findings

Persistent entropy effectively differentiates noise from features.

The method improves the reliability of topological data analysis.

Experimental results demonstrate clear separation of noise and features.

Abstract

In this paper, we derive a simple method for separating topological noise from topological features using a novel measure for comparing persistence barcodes called persistent entropy.