Natural Language Processing for Infrastructure Resilience to Natural Disasters: A Scientometric Review

New solutions are needed to enhance infrastructure resilience in response to global warming and increasing natural disasters. Recent studies demonstrate the potential of natural language processing (NLP) for mining unstructured human language during a natural disaster to assist with emergency decision-making, such as identifying infrastructure damages. NLP-based social sensing offers a way to detect disruptions and gain insights into situational awareness about critical infrastructure. Nonetheless, few studies have attempted to map the global research on applying NLP for infrastructure resilience. This paper aims to fill this gap by presenting a scientometric review of current knowledge using networks derived from the bibliographic records of the Scopus and Web of Science core collection. Specifically, this review highlights the top trends, temporal variations, significant developments, challenges, and future opportunities in NLP-based infrastructure resilience research. This study intends to improve the understanding of using NLP for infrastructure resilience against natural disasters.