基于CUK-MEANS算法的R树构建

发布时间：2018-05-07 21:23

  本文选题：K-means算法 + 传统R树　； 参考：《小型微型计算机系统》2016年02期

【摘要】：针对K-means方法的不足,提出CUK-MEANS算法,用以解决K-MEANS方法在初始值选择上的不足和对噪声点敏感的问题.传统R树索引是动态生成的,通过节点的连续插入和分裂实现整个索引的构建,这种方法会造成大量的外包矩形重叠,从而导致索引效率不高.基于CUK-MEANS算法本文进一步提出了CKR-R()算法,利用聚类技术对数据进行预处理,减少节点之间的重叠度,提高了R树的索引效率,并且采用收缩因子使节点内数据更加紧凑,提高节点的空间利用率.理论研究和实验表明所提算法具有较高的查询效率.
[Abstract]:Aiming at the shortage of K-means method, a CUK-MEANS algorithm is proposed to solve the problem of K-MEANS method in selecting initial value and being sensitive to noise points. The traditional R-tree index is dynamically generated. The whole index is constructed by the continuous insertion and splitting of nodes. This method will result in a large number of outsourced rectangular overlaps resulting in low index efficiency. Based on CUK-MEANS algorithm, this paper proposes CKR-RN) algorithm, which uses clustering technology to preprocess the data, reduces the overlap between nodes, improves the index efficiency of R-tree, and uses contraction factor to make the data of nodes more compact. Improve the space utilization of nodes. Theoretical research and experiments show that the proposed algorithm has a high query efficiency.
【作者单位】： 哈尔滨理工大学计算机科学与技术学院;
【基金】：国家自然科学基金项目(61370084)资助 黑龙江省自然科学基金项目(F201302)资助 黑龙江省教育厅科学研究项目(12541128;12531z004)资助
【分类号】：P208;TP311.13
，

本文编号：1858517