Mining And Analysis Of Spatiotemporal Characteristics Of Production Safety Accident Data In Transportation Industry

With the continuous expansion of the coverage of the transportation industry,the content of work safety accidents has become increasingly complex,and the data of work safety accidents in the transportation industry are mostly stored in the form of text,with high value but low value density.It is difficult to realize the risk prediction and prevention only by the traditional sports and manual supervision methods.How to use information technology to effectively manage the text of the traffic industry production safety accident data,and make a scientific analysis of the traffic industry production safety accident,has become a hot issue in the current research field.In view of the above problems,this paper designs and implements a text data processing method based on word vector.Through keyword classification and knowledge extraction technology,unstructured text data is processed into structured data with a variety of key attributes.Through a variety of visual analysis methods,the characteristics of production safety accidents in Beijing traffic industry are mined,and their advantages and disadvantages are compared This paper analyzes the causes and puts forward reasonable suggestions.On the basis of the above research work,the analysis results will be implemented to form a prototype system of big data platform for safety production in transportation industry,which will provide information support and scientific guidance for relevant management departments of safety production in Beijing transportation industry.The main contents of this paper are as follows:(1)This paper designs an accident text keyword classification method based on natural language processing technology,constructs a set of traffic industry related business data automatic processing process,and realizes the rapid,standardized and procedural traffic industry safety production accident data preprocessing.(2)Oracle relational database and neo4 j graph database are constructed to form the basic database that provides data services for the upper system application and visual analysis,which is conducive to the standardized management of related business data and the persistent storage of knowledge.(3)Through the construction of traditional thermal map and keyword co-occurrence map,using the method of visual analysis to mine the spatio-temporal characteristics of safety accidents in the transportation industry.This paper summarizes the main driving factors of the accident combined with the cause theme summary method,analyzes the above results and puts forward reasonable suggestions,provide information support and scientific guidance for the relevant management departments of production safety in Beijing transportation industry.(4)Based on WebGIS technology,this paper designs and implements the application development of traffic industry safety situation monitoring system,including safety situation,safety supervision,risk warning,emergency management,statistical analysis,which provides scientific decision support for emergency disposal of traffic industry in Beijing.