Analysis of Traffic Accidents in Construction Zone Reconstruction and Expansion Based on Text Mining

In the process of highway reconstruction and expansion, understanding the characteristics of traffic accidents is important for safety management and prevention. Currently, there is a relative lack of research on the impact of highway reconstruction and expansion on traffic accidents. In order to explore the characteristics of traffic accidents in the context of highway reconstruction and expansion, 372 traffic accidents on a highway are taken as the research object, and text mining and clustering model frameworks are used to identify the relevant characteristics of traffic accidents under reconstruction and expansion. Text mining is used to extract keywords and thus association rules, while strong association rules of keywords are subsequently used to construct a clustering model. The results show that in the context of highway reconstruction and expansion, accidents mainly involve cars and tend to occur in the spring of specific conditions or under specific date conditions. The results of this study provide an empirical basis and guiding direction for improving the safety management of highways in the context of reconstruction and expansion, and help to develop targeted traffic safety policies and measures to improve the safety and efficiency of road use.