Finding critical nodes in a complex network from information diffusion and Matthew effect aggregation

The identification of critical nodes is a crucial aspect in studying the spread of diseases, vaccination strategies, power grid robustness, advertisement placement, and rumor control. Consequently, this topic has become of immense interest in recent times. In the last decade, numerous methods have been proposed for identifying critical nodes, but each method has its own strengths and weaknesses, which can be attributed to the complex nature of networks and different scenarios. Therefore, it is unlikely that a single method can be applicable to all networks. To address the need for improved critical node identification in propagation scenarios, we propose a new approach called IDME (Information Diffusion and Matthew Effect aggregation). This approach is inspired by the real-world phenomenon of information diffusion and the Matthew effect. IDME simulates the dissemination of information in the real world and obtains information from multilayer neighbors, which is then aggregated using the Matthew effect. By considering its own information as well as that of its multilayer neighbors, IDME can more accurately identify critical nodes in networks while maintaining low time complexity. Experimental results on numerous real-world networks demonstrate that the IDME approach is effective in detecting critical nodes in networks and outperforms representative algorithms on most networks.