一种分布式计算的空间离群点挖掘算法

针对现有空间离群点挖掘算法无法适应大规模空间数据挖掘的需求,该文提出了一种分布式条件下的空间离群点挖掘算法。首先,该文针对集群上分布式计算和存储的特点提出使用空间填充曲线来划分数据集,加速寻找目标点的近似空间最近邻居。其次,使用信息熵的理论来定义空间离群系数,考虑到多维数据中不同属性对离群系数的影响具有差异性,该算法能够自动根据数据原有特点,计算各属性的权重;同时使用反距离权定义空间因素对离群系数的影响。最后,实验结果表明该算法在大规模的空间数据集中挖掘离群点的效率远高于传统算法,离群点的挖掘精度在90%以上。

A spatial outlier mining algorithm based on distributed computing

For the existing spatial outlier mining algorithms cannot adapt to the needs of large-scale spatial data mining,a spatial outlier mining algorithm based on distributed system was presented in this paper.Firstly,the use of space filling curve to partition the data set,and speed up the nearest neighbor of the target point were proposed.Secondly,using the theory of information entropy to define the spatial outlier factor,the effect of different attributes of multidimensional data on the outliers was taken into account and the weight of each attribute according to the original features of the data was calculated automatically;at the same time,the influence of spatial factors on the outlier factor was defined by the inverse distance weight.Lastly,experimental results showed that the efficiency of this algorithm was much higher than that of the traditional algorithm,and the accuracy of outlier mining was more than ninety percent.