For diversity measure between the mixed attribute dataset objects lost the original data characteristics, a new diversity measure method is introduced, the object mixed attribute outlier factor is constructed. And a new anomaly detection algorithm is presented based on mixed attribute dataset local density. The experimental results show that the algorithm is feasible, efficient, high detection precision, and simple parameter set.%针对混合属性数据集对象间差异性度量丢失原有数据特性的问题,引入了新的差异性度量方法,构造出对象的混合属性异常因子.在此基础上提出一种新的基于混合属性数据集的局部密度异常检测算法.实验结果表明,该算法高效可行,检测精度高,且参数设置简单.
展开▼
