障碍空间中不确定对象的组k最近邻查询方法  

Group k-Nearest Neighbor Query Method for Uncertainty Data in Obstructed Spaces

作　　者：万静[1] 唐贝贝 孙健 何云斌[1] 李松[1] WAN Jing;TANG Bei-bei;SUN Jian;HE Yun-bin;LI Song(School of Computer Science and Technology,Harbin University of Science and Technology,Harbin 150080,China)

机构地区：[1]哈尔滨理工大学计算机科学与技术学院

出　　处：《哈尔滨理工大学学报》2019年第3期29-34,共6页Journal of Harbin University of Science and Technology

基　　金：黑龙江省教育厅科学技术研究项目(12531z004)

摘　　要：针对障碍空间中不确定对象的组k最近邻查询问题,提出了Pk OGNN(probabilistic k obstructed group nearest neighbor query)查询方法。Pk OGNN查询方法主要包括4个子算法:Compadist_o(),SpatialPru(),PruInterEnt()和PkOGNN(),这些子算法分别是集总障碍距离的计算方法、空间修剪方法、根据空间修剪方法进行R树中间结点修剪、最终精炼查询方法。所提Pk OGNN查询方法通过集成有效的修剪策略以便减少Pk OGNN的搜索空间,得到正确的k GNNs。理论研究和实验结果表明,所提方法具有较好的性能。To deal with the problem of group k-nearest neighbor query method for uncertainty data in obstructed spaces,this paper presents the method of the P k OGNN(probabilistic k obstructed group nearest neighbor)query.The P k OGNN query method mainly includes four sub-algorithms:Compadist_o(),SpatialPru(),PruInterEnt()and PkOGNN(),These algorithms are respectively the calculation of the aggregate obstructed distance,the spatial pruning method,the pruning of the R-tree intermediate items according to the spatial pruning method,the final refined query method.It integrates the effective pruning methods to reduce the search space of P k OGNN and get the correct k GNNs.The theoretical research and experimental results show that the proposed method has good efficiency.

关 键 词：R树 组最近邻查询 不确定性 可视性 障碍距离 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]