基于校园无线网络的时空轨迹相似性度量  

Spatial-temporal trajectory similarity measurement based on campus wireless network

作　　者：方敏佳 刘漫丹[1] FANG Min-jia;LIU Man-dan(School of Information Science and Engineering,East China University of Science and Technology,Shanghai 200237,China)

机构地区：[1]华东理工大学信息科学与工程学院,上海200237

出　　处：《计算机工程与设计》2020年第11期3001-3008,共8页Computer Engineering and Design

摘　　要：为挖掘校园无线网络用户之间的关联性,提高用户相似性度量的准确性,针对无线网络中产生的时空轨迹数据的特征,提出一种基于最短时间距离子序列的时空轨迹相似性度量模型。同时考虑轨迹的时间参数和空间参数特征,利用最短时间距离模型求取空间相似性,采用最短时间距离子序列模型,引入连续因子体现轨迹序列特征,求取轨迹空间相似性;将时间和空间相似性汇总得到轨迹整体相似性,反映用户之间的相似性结果;利用并行滑动时间窗对用户轨迹进行划分,提高计算效率。基于真实校园无线网络数据集进行实验分析,验证了该方法在局部轨迹段和整体轨迹集中均有较好准确性。To mine the relationship between users of campus wireless network and to improve the accuracy of similarity calculation between users,according to the characteristics of trajectory data in campus wireless network,a spatial-temporal trajectory similarity measurement model based on shortest time distance sequence was proposed.The spatial and temporal characteristics of trajectories were considered together.The shortest time distance model was used to calculate the temporal similarity of trajectories.The shortest time distance subsequence model,with the introduction of continuity factor,was used to obtain the spatial similarity.The spatial and temporal similarity was aggregated to obtain the overall similarity of trajectories to reflect the similarity results between users.The parallel sliding time window was used to partition the user’s trajectory to improve the computational efficiency.Experimental analysis of real campus wireless network data shows that the proposed model has good accuracy in local trajectory segment and the overall trajectory set.

关 键 词：时空轨迹 相似性 校园无线网络 最短时间距离子序列 连续因子 

分 类 号：TP274[自动化与计算机技术—检测技术与自动化装置]