基于GPS轨迹的周期模式发现

摘要

In order to solve the problems which Data sampling frequency uncertainty, data sparseness, we propose an algorithm called GMPF (GPS Multi-Periodic Find) to detect periodic pattern in people's trajectory. It is a two-stage algorithm. At the first stage, the raw trajectory can be transformed into set of reference spots. At the second, we use a novel probabilistic measure to detect periods for each element in reference spot's set. Finally, experiments on real data set which was collected in (Microsoft Research Asia) Geolife project by 182 users in a period of over three years (from April 2007 to August 2012). The method is provably robust to incomplete observations and sparse data.%针对周期行为挖掘中面临的时空数据采样频率不确定，数据稀疏，时空数据噪声等问题，本文采用GMPF （GPS Multi-Periodic Find）算法来检测用户的周期模式。该算法首先将用户的轨迹序列转换成兴趣点集合，然后针对每个兴趣点进行周期挖掘。通过在微软亚洲研究院的Geolife项目中的182名用户4年的GPS数据上进行实验，实验证明了该方法的有效性且对数据噪声和数据稀疏不敏感。