Quantum transport senses community structure in networks.

Area of Science:

• Quantum mechanics
• Network science
• Data analysis

Background:

• Quantum time evolution involves complex interactions between wave function density and phase.
• Classical diffusion models are well-established, but quantum mechanics' wave nature is underexplored in data analysis.
• Complex networks require advanced clustering techniques to reveal underlying structures.

Purpose of the Study:

• To propose and validate a novel clustering algorithm based on quantum transport.
• To demonstrate the efficacy of Quantum Transport clustering (QTC) in analyzing complex networks.
• To provide a quantum-inspired alternative to traditional network clustering methods.

Main Methods:

• Utilizing the Laplace transform of quantum transport (QT) to map complex networks onto a circle (S¹).
• Developing the Quantum Transport clustering (QTC) algorithm to group densely connected nodes into concentrated clusters on S¹.
• Comparing QTC's performance against state-of-the-art spectral clustering for pattern discernment.

Main Results:

• QTC successfully groups related nodes into sharply concentrated clusters on S¹.
• The algorithm demonstrates power comparable to spectral clustering in identifying complex geometric patterns.
• QTC exhibits enhanced robustness, particularly with clusters of varying densities and sizes.
• The phenomenon is interpretable via an effective tight-binding model of collective node behavior.

Conclusions:

• Quantum Transport clustering (QTC) offers a powerful new approach to network analysis.
• QTC provides a robust and accurate method for discerning complex patterns in networks.
• The quantum mechanical framework offers unique advantages for data analysis and network science.