Abstract
The constant growth rate of research publications related to disaster prevention, response and resilience, however beneficial to society, has become a challenge to those working on this topic and looking for identifying the strengths, weaknesses and opportunity areas for the continuous improvement of the state of the art. This situation motivated a literature-based analysis of natural hazard and resilience using the tools of competitive technology intelligence (CTI) and artificial intelligence (AI) combined to develop charts and network maps for an efficient visualization of the current state of natural hazard and resilience research, as well as the interconnections between the different scholars and research entities, which in turn helps identifying the scientific-technological capacities of research groups, the human capital involved and potential areas of opportunity for further research. The benefits of this type of analysis may only be guaranteed if the proposed analysis of information procedure is continuously carried out, using the computational tools developed in this paper which include those of the CTI and the algorithms of multi-label machine learning designed to categorize upcoming research papers in accordance with relevant keywords into diverse predefined labels. The results of the investigation showed that a large proportion of current research on natural hazard and the resilience of affected infrastructure is carried out by researchers in China, the USA and New Zealand, and that the main focus of the published research results on this problem is on the characterization and quantification of the hazards, the physical, economic and social vulnerabilities and resilience.