Factors correlation mining on maritime accidents database using association rule learning algorithm

Abstract

Maritime safety is of paramount significance for marine industry since the maritime accidents may adversely affect the human, cargos, ships and the marine environment in various forms and degree of extent. The study aims to identify potential causal relationships among the many factors that play a role in maritime accidents. Correspondingly, association rule learning is selected as analysis approach, because of its utility in obtaining association rules through data mining on maritime accidents data. Based on the analysis of association rule learning, this study designs the association rules learning procedure of maritime accidents and establishes the association rule learning model of maritime accidents. The novelty of this study is to present a different perspective during maritime accident analysis in which potential causal relationships among the many factors are revealed. Association rule learning of maritime accidents data is carried out based on the Apriori algorithm, and the strong association rules among the causal factors of the accident are generated. The study then analyzed the generated strong association rules to find the potential relationship among the causal factors, and puts forward the coping strategies to prevent similar maritime accidents occurrence.