Abstract: When there is a railway accident, analysis of the cause of the accident is an important work in the process of accident investigation.In order to solve the problem of lack of intelligence in accident analysis methods at present, based on the parsing of railway accident big data, a design scheme of collaborative analysis system of railway safety information and technical regulations was proposed. The method and tools of Chinese word segmentation and full-text retrieval in natural language processing technology were used to establish the correlation between accident profile and technical regulations, thus improve the accuracy and intelligence of accident cause analysis.The test results of simulation data show that the developed system can reduce the work intensity of accident analysts to a large extent, and has a good prospect of application and promotion.