【摘要】 随着信息技术的快速发展,数据资源的结构越来越复杂,离群点挖掘受到越来越多人关注.基于高斯核函数,考虑数据对象的k个最近邻居,反向k近邻居和共享最近邻居三种邻居关系,估计数据对象的密度,提出了一种基于高斯核函数的局部离群点检测算法.该算法通过KNN图存储每个数据对象的最近邻,包括k最近邻,反向k近邻和共享最近邻,构成数据对象的邻居集合S;通过核密度估计KDE方法估计数据对象的密度;通过相对密度离群因子RDOF来估计数据对象偏离邻域的程度,进而判定数据对象是否为离群点,并在真实和合成的数据集上证明了该算法的有效性.更多还原

【Abstract】 With the rapid development of information technology,the structure of data resources is becoming more and more complex,and outlier mining is attracting more and more attention. Based on Gaussian kernel function,this paper considered three neighbors: k nearest neighbors,reverse k neighbors and shared nearest neighbors. A local outlier detection algorithm based on Gaussian kernel function was proposed. The algorithm stores the nearest neighbors of each data object through KNN maps,including k-nearest neighbors,reverse k neighbors,and shared nearest neighbors,forming a core neighbor set S; the estimating KDE by using kernel density with Gaussian kernel function kernel. The method estimates the density of the data object. The relative density outlier factor RDOF was used to estimate the extent to which the data object deviates from the neighborhood,it is determined whether the data was an outlier. Real and synthetic datasets prove that the algorithm was effective.更多还原