Large Language Modelsfor Injury Surveillance: Evidence from E-Bike Incidents
arXiv:2506.04654v3 Announce Type: replace Abstract: This study analyzes e-bike-related injury incidents using two complementary data sources: the Consumer Product Safety Risk Management System (CPSRMS) and the National Electronic Injury Surveillance System (NEISS). We develop an LLM-assisted framework that converts unstructured incident narratives into structured variables, notably human-, equipment-, and environment-related contributing factors, for use alongside demographic, locational, tempor
阅读原始来源来源记录
项目分析
暂无五维分析,可先查看上方来源资料。