Construction And Application For Ontology Framework Of Railway Electrical Accident Based On Knowledge Graph

Railway electrical accidents occur frequently,but existing accident records and related information are stored as unstructured data,which makes it more difficult to obtain correlations between data.Knowledge graph is the key technology to acquire knowledge and build knowledge base in the era of big data.It not only integrates scattered data,but also improves the accuracy of data mining.In order to explore the hidden information between railway electrical accidents and integrate seemingly independent data into the knowledge base,exploring the semi-automated construction process of railway electrical accident ontology framework based on knowledge graph and applying it.The railway electrical accidents recorded by the Railway Bureau from 2010 to 2018 are the main data,and these accidents are stored in unstructured documents.In order to convert unstructured data into structured data,the BiLSTM-CRF model is used to identify the time,train,location and type of content of the accident,and other structural content is manually analyzed.In order to contact the accident,the accident is classified from the faulty equipment and the type of accident.Through a variety of classifiers for experimental comparison,the classifier with the best classification effect is selected to classify the accidents.The processed structured data and the classified results are organized into a CSV file and imported into the graph database.The knowledge graph can not only analyze and diagnose faulty equipment and accident types of railway electrical accidents,but also help us predict trends and changes in these accidents.At the same time,the processed structured data and the classified results are edited using the Protege tool,and the OWL is used as the description language to manually construct the railway electrical accident ontology.The constructed ontology can be used for the query and reasoning operation.In the future research,this ontology can be used for application,and on the basis of this,ontology expansion can be performed.In order to facilitate the user to view and operate the knowledge graph of railway electrical accident,the system is used to realize the knowledge graph display and query function.Through this system,users can discover the linkage and the hidden information between accidents,and also obtain the required information conveniently and quickly.This paper takes the railway electrical accidents in recent years as the main data,and uses graph database to realize the construction of knowledge graph.Drawing on the seven-step method,the OWL language is used to describe the ontology,and the Protege tool is used to manually construct the accident ontology framework.The knowledge graph visualization and search functions are implemented using the Flask framework.After testing,the application system works well.