Similarities on Graphs: Kernels versus Proximity Measures

TL;DR

This paper analytically compares kernels and proximity measures on graphs to understand their mathematical properties, aiding in selecting appropriate similarity measures for data analysis tasks.

Contribution

It provides a detailed analytical comparison of graph kernels and proximity measures, highlighting their mathematical differences and potential applications.

Findings

Different kernels exhibit distinct mathematical properties.

Proximity measures can be characterized by their distance properties.

Insights can guide the choice of similarity measures in graph analysis.

Abstract

We analytically study proximity and distance properties of various kernels and similarity measures on graphs. This helps to understand the mathematical nature of such measures and can potentially be useful for recommending the adoption of specific similarity measures in data analysis.