Implementation of Railway Accident Judgment Criteria Optimization Based on Data Mining and Digital Programming Technology

The project team investigated the current situation of research at home and abroad, and made preliminary preparations for relevant theories and technologies, and then took the EU railway accident data as the main line to complete the whole process of research and analysis of data collection, data cleaning, data processing, data modeling, data analysis and data visualization. The eu railway accident alone and annual report for the translation and sorting, the use of digital programming technology to solve the railway accident in the process of too refinement, extracted data is not enough for data analysis or difficult to rationalize the data analysis, the original literal data streamlined classification and the cause of the accident according to the categories group small class coding. Tosuch as Python language and MATLAB were introduced to conduct accident cause chain analysis to improve the effectiveness of data analysis. On this basis, the railway accident grade is quantified based on the entropy weight method to realize the optimization of the railway accident evaluation standard in the quantitative analysis. The project team put forward the establishment of a data banking structure for the innovation of railway accident data asset management, and put forward relevant suggestions on the subsequent railway accident prediction and early warning mechanism. The research results of the project have high application value to academic research, the government, railway enterprises, railway staff and other relevant parties.