摘要
R-Tree允许兄弟节点之间的相互重叠,具有多路查找的特点,而Hilbert R-Tree也不能有效降低子空间的相互重叠,直接影响查询效率。提出了一种基于混合聚类的空间索引算法,将K-means和K中心点引入索引结构,改变了经典K-means算法对初始聚类中心的随机选取,减少了叶节点的MBR面积和各个子空间的重叠。通过实验表明,该算法具有更快的响应速度和查询效率。
The R-Tree spatial index structure was analyzed.There are overlap between brothers nodes and multi-path in search ,and Hilbert R-Tree can not effectively reduce the overlap, which is a direct impact on query efficiency. Based on hybrid spatial clustering algorithms, a spatial index algorithm used K-means algorithm and K-center algorithm is proposed, which improve the random choice of the initial centrists in the classic K-means algorithm and decrease the leaf nodes MBR area and overlap between interior nodes. Experiments show that the algorithm has the faster response speed and the higher query efficiency.
作者
韩秋英
马骏
张少辉
HAN Qiu-ying1, MA Jun1,2, ZHANG Shao-hui3 (1.College of Computer and Information Engineering, Henan University, Kaifeng 475004,China;2.Institute of Data and Knowledge Engineering, Henan University,Kaifeng 475004,China;3.Department of Computer Science, Zhoukou Normal University,Zhoukou 466000,China)
出处
《电脑知识与技术》
2009年第12Z期10047-10048,10056,共3页
Computer Knowledge and Technology
