Uses a BERT/Soft Lexicon/CRF-style information extraction approach to structure Chinese chemical accident reports and extract accident information and risk factors. It is relevant to Process Safety Management because many companies have large libraries of unstructured incident reports; this type of NLP method can turn narrative investigations into standardized data for root-cause learning, trend analysis, and prevention planning.
AUTHORS
Xi Luo; Xiayuan Feng; Xu Ji; Yagu Dang; L. Zhou; K. Bi; Y. Dai
CITATIONS
X. Luo, X. Feng, X. Ji, Y. Dang, L. Zhou, K. Bi, and Y. Dai, "Extraction and analysis of risk factors from Chinese Chemical Accident Reports," Chinese Journal of Chemical Engineering, vol. 61, pp. 68–81, 2023, doi: 10.1016/j.cjche.2023.02.026.
