Dynamic Railway Derailment Risk Analysis with Text-Data-Based Bayesian Network

In recent years, transportation system safety analysis has become increasingly challenging and highly demanding. Unstructured data contain sufficient information from which inherent interactions can be extracted. Determining how to process fuse a large amount of unstructured is task. this paper, we propose text-based Bayesian network (TBN) method establish (BN) based on text records, where the BN’s arcs are obtained barrier relationships identified by graphical model its prior probabilities stem fault trees. The comparative experimental results illustrate that in TBN efficient. precision, recall F-measure 8.64%, 10.70% 9.84% higher, respectively, than most frequent (MF) result. Moreover, compared traditional BN, whose frequently acquired experts, proposed BN have high confidence. train derailment accident case study show with changes potentials barriers, generates quantitative reveals critical risks accidents. Additionally, work demonstrates relevant nonlinear improve assessment results. Therefore, data, potential identify high-risk guide managers enhance these barriers.