Quantifying layer similarity in multiplex networks: a systematic study

Area of Science:

• Network Science
• Data Analysis

Background:

• Multiplex networks, comprising multiple layers of interconnected nodes, are prevalent in various systems.
• Characterizing relationships between these layers is essential for understanding network properties and dynamics.

Purpose of the Study:

• To provide a systematic taxonomy of methods for computing layer similarities in multiplex networks.
• To experimentally evaluate existing and novel approaches for layer comparison.
• To offer practical guidelines for selecting and applying appropriate comparison methods.

Main Methods:

• Development of a comprehensive taxonomy categorizing layer similarity computation techniques.
• Experimental evaluation of different similarity metrics across diverse multiplex network structures.
• Analysis of the performance and applicability of various approaches.

Main Results:

• The study categorizes and extends existing layer similarity methods.
• Experimental results provide insights into the effectiveness of different approaches under various conditions.
• Practical guidelines are established for researchers and practitioners.

Conclusions:

• A structured framework for understanding and applying layer similarity computations in multiplex networks is presented.
• The findings facilitate more accurate characterization and analysis of complex multiplex systems.
• This work enhances the ability to study static and dynamic processes within multiplex networks.