一种基于Hadoop的高效空间拓扑关系判定方法

互联网数据规模增长迅速,作为判定位置相关性的空间分析方法在互联网数据分析中得到广泛应用,空间拓扑关系判定是空间分析算法的重要基础。考虑到互联网数据的海量规模,采用传统方法的空间拓扑关系判定已远超出当前系统的处理能力。文中以分层法和二分查找法为基础提出了一种判定海量地理坐标与给定区域空间拓扑关系的高效判定算法,并结合Hadoop/Hive平台对算法进行优化。实验结果表明该方法可高效实现对海量空间数据的拓扑关系判定,并在算法精度上具有与ESRI传统算法相媲美的结果。

A Hadoop-based effective determination method for spatial topological relations

As the scale of Internet data growing rapidly, the spatial analysis methods of determining the position correlation has been widely used in the Internet Data Analysis, determining topological relations is an important foundation for spatial analysis algorithms. Considering the massive size of Internet data, traditional processing methods of which has gone far beyond the capabilities of current system. Basing on the hierarchical method and binary search method, the paper designed an effective determination algorithm of the spatial topological relations between large sets of geographic coordinate and given regions, and optimized it based on Hadoop/Hive platform. The experimental results showed that the method can determine the topological relations of massive spatial data effectively, and it has the equal accuracy comparing to ESRI traditional algorithm.