基于Delaunay三角网的空间离群挖掘

空间离群是指非空间属性与其空间邻居显著不同的空间对象。空间数据的特殊性决定了空间离群挖掘需要充分考虑空间数据的特点,才能挖掘出有现实意义的离群。本文对现有主要的空间数据离群挖掘算法进行了研究分析,针对k-邻域法确定空间邻域的缺点,基于Delaunay三角网在表达空间邻近关系的有效性,通过构建Delaunay三角网确定空间邻域并生成空间权重矩阵,据此提出了基于Delaunay三角网的空间离群挖掘算法DT_SOF,并以实际生态地球化学数据进行实验检验。结果表明,算法具有较低的用户依赖性,能准确挖掘空间离群。

Spatial Outlier Detection Based on Delaunay Triangulation

Spatial outliers represent objects whose non-spatial attribute values are significantly different from the values of theirs spatial neighborhoods,in order to mining the valuable outliers,spatial outliers mining should consider the special characteristics of spatial data.The major drawbacks of existing methods are that spatial characteristics aren't considered fully and they use k-neighborhood method to define neighborhood which depends on a priori given parameters.So we propose a new neighborhood-defined algorithm and compute the spatial weighted matrix which is based on Delaunay Triangulation.In virtue of that,the spatial outliers detection based on Delaunay Triangulation is proposed.In addition,using a real-world ecological geochemical dataset,we demonstrate that our approach is efficient and has lower human dependence.