The frequency and severity of catastrophic disasters have escalated over the past decade, contributing to significant global economic losses, which has underscored the urgent need to leverage advanced knowledge and innovative technologies to strengthen disaster management. Scholars and practitioners alike acknowledge that the application of novel technologies and advanced methodological approaches can substantially mitigate the adverse impacts of disasters on life and property. Gaps are evident in relation to trend analysis, analytical methods employed, nature and origin of data sources (primary vs. secondary), prevailing shortcomings, and areas for future research. This bibliometric study systematically analyzes global research trends in advanced data analytics for disaster management from 2006 to 2025, examining publication patterns, methodological evolution, international collaboration networks, and research gaps. We retrieved 925 documents from Scopus and employed Bibliometrix, PyBibX, and network analysis tools to conduct citation analysis, keyword co-occurrence mapping, and thematic evolution assessment. Results shows United States as the hub of studies using advanced methods in disaster management with its publications reporting highest citations followed by China and India respectively. Social media has surfaced as the most prevalent area of research, followed by Machine learning and Deep learning. Emerging methods such as Multi-Task Learning, Disaster Tweets, and Infrastructure Damage Detection exhibit a temporal concentration in 2023–2024, indicating these topics are situated at the forefront of recent scholarly inquiry. The basic themes quadrant (bottom-right) includes Deep Learning, Disaster Management, and Crisis Management while emerging/declining quadrant (bottom-left), topics such as BERT, Transformers, Social Computing, Natural Language Processing (NLP), Twitter, and Topic Modeling exhibit low centrality and density. Floods remain one of the prominent focuses of disaster literature using NLP and Deep Learning approaches primarily addressing the SDG11, whereas the ensemble approaches are underutilized. • Leverage advanced knowledge to strengthen disaster management research and practice.. • United States as the hub of studies using advanced methods in disaster management • Social media, Machine learning, Deep learning are most prevalent methods. • Multi-Task Learning, Disaster Tweets are emerging approaches in Disaster Management. • SDG11 is the focus while other SDGs need for concentration in future studies.
Maghelal et al. (Sun,) studied this question.