Using Structural Topic Modeling to Explore Aviation Safety Reporting System Data

The Aviation Safety Reporting System includes over a million confidential reports describing safety incidents. Natural language processing techniques allow for relatively rapid and largely automated analysis of large collections of text data. Meaningful interpretation of the results and further investigations by subject matter experts can follow. This article describes the application of structural topic modeling to Aviation Safety Reporting System data. Results reveal that the application is able to identify known issues. The method also has the potential to identify previously unknown connections that may warrant further, more manual, study. Results reported here highlight the importance of fuel pump, tank, and landing gear issues and the relative insignificance of smoke and fire issues for private aircraft. The results also uncovered evidence of the prominence of the Quiet Bridge Visual and Tip Toe Visual approach paths at San Francisco International Airport in safety incident reports. Keywords-aviation safety; machine learning; structural topic modeling; natural language processing;