Abstract
This study analyzed the causes and trends of chemical accidents that occurred in Korea from 2014 to 2022. In particular, by comprehensively considering human casualties, environmental damage, and property damage, the relative risks and accident trends of each type of accident were identified. Utilizing the relative risk, the most dangerous accident type for each type of accident were derived. And, LDA topic modeling and network analysis were used to analyze accident trends for fire accidents identified as the most dangerous accident types. Results showed that the most dangerous accident type in terms of human casualties was explosion accidents, and the type of accident with a large impact on the environment and property was analyzed as fire accidents. As a result of LDA topic modeling, chemical fire accidents were broadly classified into four major topics. It is possible to analyze the complex causes and patterns of chemical accidents through methods using LDA topic modeling and network analysis, and it is possible to effectively establish prevention and response strategies for chemical accidents through the methods used.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
