Comparative evaluation of community-aware centrality measures

Influential nodes play a critical role in boosting or curbing spreading phenomena complex networks. Numerous centrality measures have been proposed for identifying and ranking the according to their importance. Classical rely on various local global properties of nodes. They do not take into account network community structure. Recently, growing number researches shifted community-aware measures. Indeed, it is ubiquitous feature vast majority real-world In literature, focus designing However, up now, there no systematic evaluation effectiveness. This study fills this gap. It allows answering which measure should be used practical situations. We investigate seven influential an epidemic process scenario using Susceptible–Infected–Recovered model set fifteen Results show that generally, correlation between low. Furthermore, multiple-spreader problem, when resources are available, targeting distant hubs Modularity Vitality more effective. with limited resources, diffusion expands better through bridges, especially networks medium strong