不均匀模糊空间对象的分层次co-location模式挖掘方法
发布时间：2019-06-10 18:21
【摘要】：针对现有的co-location模式挖掘算法无法有效处理不均匀分布空间对象的问题,提出一种不均匀模糊空间对象的分层次co-location模式挖掘方法。首先提出一种不均匀数据集的生成方法;然后对不均匀分布的数据集进行层次划分,使每个区域具有均匀的空间分布;再基于改进的PO_RI_PC算法对划分后的模糊对象进行空间数据挖掘。该方法基于距离变化系数构建每个子区域的邻域关系图,进而完成区域融合,实现co-location模式挖掘。实验结果表明,与传统方法相比,所提方法的执行效率更高,随实例个数和不均匀度的变化获得的co-location集个数更多,同比情况下平均提高约25%,获得了更精确的挖掘结果。
[Abstract]:In order to solve the problem that the existing co-location pattern mining algorithms can not effectively deal with uneven distributed spatial objects, a hierarchical co-location pattern mining method for inhomogeneous fuzzy spatial objects is proposed. Firstly, a generation method of uneven data set is proposed, and then the uneven distribution data set is divided into layers so that each region has uniform spatial distribution. Then based on the improved PO_RI_PC algorithm, the spatial data mining of the divided fuzzy objects is carried out. Based on the distance variation coefficient, the neighborhood diagram of each sub-region is constructed, and then the region fusion is completed to realize co-location pattern mining. The experimental results show that compared with the traditional method, the proposed method has higher execution efficiency, and more co-location sets are obtained with the change of the number of examples and inhomogeneity, and the average increase is about 25% compared with the same period last year, and more accurate mining results are obtained.
【作者单位】： 安徽师范大学数学计算机科学学院;安徽师范大学国土资源与旅游学院;
【基金】：国家自然科学基金资助项目(61370050,61572036) 安徽省自然科学基金资助项目(1508085QF134) 安徽师范大学创新基金资助项目(2016XJJ074)~~
【分类号】：TP311.13
，
本文编号：2496638
[Abstract]:In order to solve the problem that the existing co-location pattern mining algorithms can not effectively deal with uneven distributed spatial objects, a hierarchical co-location pattern mining method for inhomogeneous fuzzy spatial objects is proposed. Firstly, a generation method of uneven data set is proposed, and then the uneven distribution data set is divided into layers so that each region has uniform spatial distribution. Then based on the improved PO_RI_PC algorithm, the spatial data mining of the divided fuzzy objects is carried out. Based on the distance variation coefficient, the neighborhood diagram of each sub-region is constructed, and then the region fusion is completed to realize co-location pattern mining. The experimental results show that compared with the traditional method, the proposed method has higher execution efficiency, and more co-location sets are obtained with the change of the number of examples and inhomogeneity, and the average increase is about 25% compared with the same period last year, and more accurate mining results are obtained.
【作者单位】： 安徽师范大学数学计算机科学学院;安徽师范大学国土资源与旅游学院;
【基金】：国家自然科学基金资助项目(61370050,61572036) 安徽省自然科学基金资助项目(1508085QF134) 安徽师范大学创新基金资助项目(2016XJJ074)~~
【分类号】：TP311.13
，
本文编号：2496638
本文链接：https://www.wllwen.com/kejilunwen/ruanjiangongchenglunwen/2496638.html
