An empirical research of critical incident of earthquake disaster based on credible association mining

摘要

Based on the credible association rule (CAR) and its maximum clique mining algorithm and combined with the composition of the evolution of risk events, this paper analyzes the data of 9 earthquakes which bring greater impact to the world than any others in recent 4 years, simultaneously, definites 14 typical critical incidents of earthquake disaster. According to the results of data mining, this paper represents the credible association relationships and the probabilities of these critical incidents in different areas. Moreover, the paper discusses the results of maximum clique to prove the exactness of these credible association relationships, and reflects the effectiveness of the CAR and its mining algorithm based on maximum clique to a certain degree. As a result, it could be used to broadcast the occurring probability of the relevant critical incidents under the situation of the earthquake, in order to achieve the purpose of forecasting and early warning.