基于改进H-mine算法的内河航标事故关联规则挖掘  

Association rules mining for inland navigation mark accidents based on the improved H-mine algorithm

作　　者：田永忠 初秀民[2,3] 倪汉杰 蒋仲廉 初晓 TIAN Yong-zhong;CHU Xiu-min;NI Han-jie;JIANG Zhong-lian;CHU Xiao(Jiujiang Waterway Division of Yangtze River,Jiujiang 332005,China;National Engineering Research Center for Water Transport Safety,Wuhan University of Technology,Wuhan 430063,China;Intelligent Transportation System Research Center,Wuhan University of Technology,Wuhan 430063,China)

机构地区：[1]长江九江航道处,九江332005 [2]武汉理工大学国家水运安全工程技术研究中心,武汉430063 [3]武汉理工大学智能交通系统研究中心,武汉430063

出　　处：《水道港口》2022年第4期536-542,共7页Journal of Waterway and Harbor

基　　金：国家重点研发计划项目(2018YFB1600400);国家自然科学基金项目(52071250,51709220)。

摘　　要：针对内河航标事故致因关联规则数据挖掘问题,根据内河航标事故数据集事务数较少但多维多层的特点,设置算法遍历约束改进了H-mine算法,实验测试表明:改进算法可提高运算速度20%以上。同时,以长江干线航道武汉段航标事故数据为算例,开展了关联规则挖掘研究,揭示了长江干线武汉段航标事故特点与成因,验证了算法的有效性,并提出了预防航标事故的建议。The inland waterway navigation mark accident dataset has a small number of transactions but a large number of dimensions and layers.To resolve the problem of association rules mining by inland navigation mark accidents,the classic H-mine algorithm was improved with traversal constraints.Test results show that the computational efficiency is increased by more than 20%.A case study of associate rules mining has been carried out in the Wuhan channel of the Yangtze River,China.The characteristics and causes of navigation mark accidents were revealed.The effectiveness of associate rules mining algorithm was also verified,on basis of which accident prevention suggestions were proposed for aids to navigation.

关 键 词：改进H-mine算法 关联规则 航标事故数据 内河航标 

分 类 号：U675.85[交通运输工程—船舶及航道工程]