海事事故的聚类与关联规则
|
摘要
为进一步分析海事事故发生的潜在原因,保障船舶航行安全,在对比各海事局事故数据质量的基础上,以浙江海域海事事故数据库作为研究对象,通过对海上船舶风险及其影响因子进行识别,以事故类型为聚类中心,并与关联规则Apriori算法进行融合,以较高的挖掘精度实现对海事事故数据的致因分析。挖掘试验结果表明:基于k-medoids和Apriori组合的挖掘算法在提升度和挖掘精度上均优于传统的Apriori算法。通过对逐条关联规则进行解释,分析出浙江海域海事事故的特征及致因,提出预防海事事故的建议和对策,保障浙江海域船舶的航行安全。
The case study concerning the accidents in Zhejiang waters is introduced. The risk of ship accidents in the area and the influencing factors associated are identified. The accidents are clustered according to the types of accidents. Integration of the Apriori algorithm improves the accident classification and cause analysis. The results of mining experiments show that the mining algorithm based on the combination of k-medoids and Apriori is superior to the traditional Apriori algorithm in both promotion value and mining accuracy. The characteristics and causes of maritime accidents in Zhejiang sea area are studied individually through analyzing the association rules. The measures of improving maritime safety in Zhejiang waters are proposed.
The case study concerning the accidents in Zhejiang waters is introduced. The risk of ship accidents in the area and the influencing factors associated are identified. The accidents are clustered according to the types of accidents. Integration of the Apriori algorithm improves the accident classification and cause analysis. The results of mining experiments show that the mining algorithm based on the combination of k-medoids and Apriori is superior to the traditional Apriori algorithm in both promotion value and mining accuracy. The characteristics and causes of maritime accidents in Zhejiang sea area are studied individually through analyzing the association rules. The measures of improving maritime safety in Zhejiang waters are proposed.
引文
[1] LI Hongxi. Study on the Data-Mining Based Mechanism for the Occurrence of Ship Collision Accidents Caused by Human Factors. Information Technology for Manufacturing Systems V[C]. Applied Mechanics and Materials 686.
[2] DX,DS. An Application of Data Mining Tools for the Study of Shipping Safety Inrestricted Waters[J]. Safety Science,2011(49):192-197.
[3]王凤武,吴兆麟,齐壮.基于粗糙集理论的开航安全性评价结果分析[J].中国航海,2013,36(2):61-66.
[4]李波.天津港复式航道通航安全的灰色模糊预评价[D].大连海事大学,2010.
[5]杨国宾.基于遗传算法的神经网络用于确定船舶碰撞危险度的研究[D].天津:河北工业大学,2006.
[6] GOERLANDT,GOITE H,OSIRIS A. An Analysis of Wintertime Navigational Accidents in the Northern Baltic Sea[J]. Safety Science,2017,92:66-84.
[7]刘正江,吴兆麟.基于船舶碰撞事故调查报告的人的因素数据挖掘[J].中国航海,2004(2):1-6.
[8]张晓辉,刘正江,吴兆麟.基于水上交通安全事故基础数据的模式发现[J].大连海事大学学报,2011(2):39-42.
[9]张晓辉.云理论和数据挖掘在水上安全分析中的应用[D].大连:大连海事大学,2011.
[10]黄常海,高德毅,胡甚平,等.基于Apriori算法的船舶交通事故关联规则分析[J].上海海事大学学报,2014,35(3):18-22.
[11]马建斌,陈红玉.基于数据挖掘的船舶交通事故关联分析[J].船海工程,2016,45(A01):266-270.
[12]陈燕,李桃迎.数据挖掘与聚类分析[M]大连:大连海事大学出版社,2012.
[2] DX,DS. An Application of Data Mining Tools for the Study of Shipping Safety Inrestricted Waters[J]. Safety Science,2011(49):192-197.
[3]王凤武,吴兆麟,齐壮.基于粗糙集理论的开航安全性评价结果分析[J].中国航海,2013,36(2):61-66.
[4]李波.天津港复式航道通航安全的灰色模糊预评价[D].大连海事大学,2010.
[5]杨国宾.基于遗传算法的神经网络用于确定船舶碰撞危险度的研究[D].天津:河北工业大学,2006.
[6] GOERLANDT,GOITE H,OSIRIS A. An Analysis of Wintertime Navigational Accidents in the Northern Baltic Sea[J]. Safety Science,2017,92:66-84.
[7]刘正江,吴兆麟.基于船舶碰撞事故调查报告的人的因素数据挖掘[J].中国航海,2004(2):1-6.
[8]张晓辉,刘正江,吴兆麟.基于水上交通安全事故基础数据的模式发现[J].大连海事大学学报,2011(2):39-42.
[9]张晓辉.云理论和数据挖掘在水上安全分析中的应用[D].大连:大连海事大学,2011.
[10]黄常海,高德毅,胡甚平,等.基于Apriori算法的船舶交通事故关联规则分析[J].上海海事大学学报,2014,35(3):18-22.
[11]马建斌,陈红玉.基于数据挖掘的船舶交通事故关联分析[J].船海工程,2016,45(A01):266-270.
[12]陈燕,李桃迎.数据挖掘与聚类分析[M]大连:大连海事大学出版社,2012.