摘要
为快速直观的分析某一区域内的港口竞争形势,提出了一种基于广义序列模式(GSP)算法的港口竞争性分析方法,从AIS数据中提取船舶航行的靠港序列集,使用优化后的DBSCAN算法将其聚类,进而使用GSP算法挖掘出频繁序列并找出所有的竞争关系和竞争强度。使用2020年第三季度的AIS数据进行实验,挖掘出了国内港口排名前二十的竞争关系,结果显示部分港口面临着严峻的竞争局势,表明上述方法具有一定的应用价值。
To quickly and intuitively analyze the port competition situation in a certain region,a port competitiveness analysis method based on generalized sequence pattern(GSP)algorithm was proposed.The port sequence set of ships sailing was extracted from AIS data,and the optimized DBSCAN algorithm was used to cluster them.Then GSP algorithm was used to mine frequent sequences and found out all competitive relations and competitive intensity.The AIS data of the third quarter of 2020 were used to conduct the experiment,and the top 20 competitive relations among domestic ports were mined.The results indicate that some ports are facing severe competitive situations.It shows that the research method has certain application value.
作者
黄子硕
胡勤友
孙洋
HUANG Zi-shuo;HU Qin-you;SUN Yang(College of Merchant Shipping,Shanghai Maritime University,Shanghai 201306,China)
出处
《计算机仿真》
北大核心
2023年第3期153-156,共4页
Computer Simulation
基金
国家自然科学基金资助项目(71804059)
上海市科学技术委员会社会发展领域重大项目(18DZ1206300)。
关键词
港口竞争
聚类
关联规则
Port competition
Clustering
Association rules