The Research On Risk Management Of Civil Aviation Batch Unsafe Events Based On Text Mining Technology

With the rapid development of the economy,people’s travel needs are getting stronger and stronger,and the requirements for the quality of journeys are becoming higher and higher.The civil aviation is growing at a rate that exceeds people’s imagination.At meanwhile,some recent air crashes,especially the B737-MAX accidents,have constantly reminded people that civil aviation safety risk cannot be ignored.At present,civil aviation safety risk management mainly depends on the statistical analysis of structured data.The use of textual data is only limited to accident investigation reports,and there is a lack of research on other textual data.The research content also rarely involves the analysis of the correlation between events and the estimation of future security risk trends.China Civil Aviation has accumulated a large number of texts of unsafe events for many years.If fully utilized,it will play an important positive role in the safety risk management of China Civil Aviation.This article mainly uses the theory of project management to treat a single unsafe event as a project,a batch of unsafe events as a project set.And then combines the theory and methods of project risk management with text mining technology to apply to unsafe event projects.Complete risk identification,analysis,response and monitoring,and build a complete civil aviation safety risk management and control system.This article first completes the risk analysis through the following three steps and provides a basis for risk management and control.First,by analyzing the unsafe events from the basic dimensions such as event types and aircraft types,an overview of civil aviation safety risks is obtained.Second,use the matrix analysis method to calculate the risk probability of each type of unsafe event,assess the risk impact,and quantitatively calculate the comprehensive risk value.Third,use topic-based text mining technology to process batches of unsafe events,extract risk topics,and quantitatively analyze the risk characteristics of the unsafe event itemset.Then,based on the risk analysis results,propose risk management measures,and develop a risk monitoring process.The research uses a large amount of textual data,combines project management theory,project risk management theory,and forefront information analysis technology,and provides a new solution for civil aviation safety risk management and control.It also has certain reference value for risk management and control in other industries.The main deficiency of this study is that the data source is single,based only on unsafe events data,and does not involve aircraft maintenance information,crew information,etc.The risk analysis is not multidimensional.Establishing a big data platform,integrating multi-dimensional data,and comprehensively analyzing risks will be an important development direction for future civil aviation safety risk management and control.