Compressing the Chronology of a Temporal Network with Graph Commutators

Area of Science:

• Network science
• Computational epidemiology
• Data analysis

Background:

• Temporal networks are often studied using discrete time snapshots.
• Aggregating these snapshots can simplify analysis but may alter dynamics.
• Efficient methods are needed to compress temporal network data without losing critical information.

Purpose of the Study:

• To develop a method for compressing temporal network chronologies.
• To assess the impact of snapshot aggregation on network dynamics.
• To validate the method using epidemic modeling on real-world data.

Main Methods:

• Proposing a novel method to progressively combine snapshot pairs.
• Utilizing matrix commutators to quantify the dynamical effect of aggregation.
• Applying the compression technique to epidemic modeling simulations.

Main Results:

• The proposed method allows for significant compression of temporal network data.
• Snapshot aggregation, when guided by minimal dynamical effect, preserves epidemic dynamics.
• The approach was successfully demonstrated on contact tracing datasets.

Conclusions:

• Temporal network analysis can benefit from intelligent snapshot aggregation.
• The proposed method offers a balance between data compression and dynamical fidelity.
• This technique is valuable for computational epidemiology and large-scale network studies.