The locatability of Pearson algorithm for multi-source location in complex networks.

Area of Science:

• Network Science
• Information Propagation
• Complex Systems Analysis

Background:

• Understanding information or disease spread in complex networks is crucial.
• Existing source localization methods often require detailed knowledge of propagation dynamics.
• Sparse observational data presents a significant challenge in real-world network analysis.

Purpose of the Study:

• To develop a robust algorithm for locating multiple propagation sources in complex networks.
• To enable source localization without prior knowledge of propagation dynamics or parameters.
• To investigate the locatability and optimize observer node selection for accurate source identification.

Main Methods:

• A novel multi-source location algorithm based on sparse observations.
• Calculating node centrality using the correlation between node inform time and geodesic distance.
• Employing a greedy algorithm for optimal observer node selection to enhance locatability.

Main Results:

• The proposed algorithm demonstrates high location accuracy for any number of sources.
• It is robust and effective even with limited, sparse observational data.
• Simulations on both model and real-world networks validate the algorithm's feasibility and performance.

Conclusions:

• The developed algorithm provides an effective solution for multi-source localization in complex networks.
• It overcomes the limitation of requiring prior knowledge of propagation dynamics.
• The method offers a practical approach for identifying propagation origins in diverse network environments.