BCAN-RA-AI: A Novel Framework for Optimizing Disaster Recovery Information Dissemination in Online Social Networks

The faster and efficient propagation of information pertaining to disasters is very important for the successful recovery and emergency response tasks. Online Social Networks (OSNs) have a significant role to play when it comes to effective dissemination; however, the intricate and ever-changing nature of these structures means their ability to communicate effectively is often limited. This research proposes a novel, high-level framework, named BCAN-RA_AI, that integrates the Betweenness Centrality Analysis (BCAN) and Retrieval-Augmented Artificial Intelligence (RA-AI) methods in order to adjust the information flowing configuration within the digital networks, e.g. the)'s on OSNs. The BCAN component detects intermediate nodes with high betweenness centrality, which helps to increase the message propagation speed through various communities of the network. In parallel, the RA-AI component is enriching the process, fetching contextually relevant information, writing adaptive communication strategies and continuously changing in real-time based on the changing situation data. Experimental validation based on real disaster data shows that BCAN-RA-AI has an important effect to reduce the dissemination latency and enable better precision and audience scope. The proposed framework creates a solid basis for intelligent, data-driven disaster communication systems that will improve the efficiency of relief coordination and optimize the results of crisis management.