Sequential seeding in multilayer networks

Area of Science:

• Network Science
• Social Network Analysis
• Computational Social Science

Background:

• Multilayer networks model complex systems with diverse interactions (social, biological, etc.).
• Influence spread is a key process studied on these networks, crucial for understanding opinion dynamics and information diffusion.
• Identifying optimal 'seed' nodes is vital for maximizing influence propagation efficiency.

Purpose of the Study:

• To develop and evaluate a sequential seeding technique specifically for multilayer networks.
• To compare the performance of sequential seeding in multilayer networks against traditional methods used in single-layer networks.
• To analyze the impact of sequential seeding on influence coverage, budget efficiency, and process duration.

Main Methods:

• Development of a novel sequential seeding algorithm tailored for the complexities of multilayer networks.
• Empirical evaluation of the proposed technique on simulated or real-world multilayer network data.
• Comparative analysis against baseline seeding strategies (e.g., traditional methods on equivalent single-layer networks).

Main Results:

• Sequential seeding in multilayer networks significantly outperforms traditional approaches in terms of influence coverage.
• The proposed technique demonstrates improved seeding budget efficiency, requiring fewer initial nodes for comparable spread.
• A trade-off was observed, with sequential seeding in multilayer networks potentially extending the overall duration of the influence spreading process.

Conclusions:

• Sequential seeding is an effective strategy for optimizing influence spread in multilayer networks.
• The method offers practical benefits for budget-conscious campaigns and maximizing reach in complex systems.
• Further research could explore optimizing the temporal dynamics of sequential seeding to mitigate extended durations.