嵌入式数据库中多维离散数据检测仿真

摘要

A detection method of multi-dimensional discrete data in the embedded database is proposed based on the subspace division theory.Firstly,the high dimensional data are mapped into the low dimensional subspace in the embedded database and the dataset is divided into the disjoint subspace set.Then,the advantages and disadvantages of subspace division are judged according to the division skewness,and the attribute value of partial outlier factor in the optimal division subspace is worked out.Finally,the Euclidean distance is used as the distance function of discrete data detection,and the discrete data integrated with the minimum division boundary matrix are detected.The simulation results show that the proposed method can improve the detection precision of multi-dimensional discrete data effectively.The detection efficiency is high.%对嵌入式数据库中多维离散数据的检测,可快速有效地提取所需数据.对多维离散数据进行检测时,需要将数据集划分为不相交的子空间集合,再依据划分斜偏度判断子空间划分的优劣,完成数据检测.传统方法利用粗糙集理论对数据特征进行缺陷识别,但不能判断出划分后子空间的优劣,导致数据检测不准确.提出基于子空间划分理论的嵌入式数据库中多维离散数据检测方法.该方法先把嵌入式数据库中的高维数据投影至低维子空间,将数据集划分为不相交的子空间集合,然后依据划分斜偏度判断子空间划分的优劣,计算最优划分子空间中数据对象的局部离群因子属性值,将欧几里德距离作为嵌入式数据库离散数据检测距离函数,结合最小划分边界矩阵对离散数据进行检测.实验结果表明,所提方法能够有效提升嵌入式数据库中多维离散数据检测精度,且检测效率较高.