一种检测时空数据中重要同现模式的快速算法

同现模式反映了对象移动过程中的接触情况。快速准确地挖掘移动对象的同现模式是许多领域开展研究的基础。本文提出了一种从移动对象的历史轨迹数据集中快速检测重要同现模式的算法。重要同现模式是指发生在移动对象停留期间的同现模式。该算法首先检测出所有移动对象的经停地,然后以经停地为最小单位,检测经停地之间的同现,最后在经停地的同现中识别出最终的重要同现实例。在青海湖斑头雁迁徙轨迹数据集上进行了大量对比实验,验证了算法的效率及有效性。

A Fast Method for Detection of the Important Co-Occurrence Cases from Spatio-Temporal Dataset

Co-occurrence pattern reveals the interactions among objects. Detecting the co-occurrence cases of mobile objects quickly and accurately is the key step to do further researches in many research areas. In this paper, we propose a fast method to detect the important co-occurrence cases from spatio-temporal datasets. Important co-occurrence cases denote the co-occurrence occurs during the stay of mobile objects. The method we proposed is stay-site-based. Concretely, we firstly detect all the stay sites of mobile objects, and then consider stay site as the smallest unit to detect the spatio-temporal overlaps among stay sites.Finally, we figure out all the important co-occurrence cases from stay-site-based co-occurrence. To verify the efficiency and accuracy of out method, we have conducted a sort of experiments on the historical trajectories dataset of bar-headed gooses in Qinghai Lake area.