Quick spatial outliers detecting with random sampling

摘要: Existing Density-based outlier detecting approaches must calculate neighborhood of every object, which operation is quite time-consuming The grid-based can detect clusters or outliers with high efficiency, but the have their deficiencies We proposed new spatial approach random sampling This method adsorbs thought and extends density-based to quickly remove clustering points, then identify It quicker than based on queries has higher precision experimental results show that our outperforms existing methods query.