基于频繁项集的海上目标编组挖掘

Mining Marine Grouped Targets Based on Frequent Itemsets

从海量海上目标数据中挖掘出编组信息,并评估其任务能力,对海上目标身份刻画有着重要意义。探讨了轨迹相似及轨迹聚类理论在海上目标编组挖掘中的应用模式,提出一种基于频繁项集的目标编组快速挖掘算法,并提出基于海量数据的海上目标编组模型,可以自动挖掘并给出不同置信度参数下的潜在目标编组列表。基于船舶AIS数据,实验验证了该方法的有效性,能够从目标轨迹大数据中得到海上目标的典型编组。

It is of great significance to make characterization of the identity of marine targets by mining grouped information from massive marine target data and evaluate their mission capability.Firstly,the application pattern of track similarity theory and track clustering theory in grouped target mining is discussed,and then a fast mining algorithm for target grouped based on frequent itemsets is proposed,which can automatically mine and give a list of potential target groups by different confidence parameters.The effectiveness of this method is verified by experiments based on ship AIS data,and the typical grouped marine targets can be obtained from a large number of target track data.