摘要
针对基于欧式距离的最近邻居的缺失值估计算法的不足,提出了一种基于马氏距离的估计算法来估计飞行数据集中的缺失数据。该算法通过飞行数据之间的马氏距离来选择最近邻居数据,并将已得到的估计值应用到后续的估计过程中,然后采用信息熵来计算最近邻居的加权系数,得到缺失数据的估计值。仿真结果表明该算法优于基于欧式距离的最近邻居缺失值处理算法,是一种有效的飞行数据缺失值估计方法。
A estimated method based on Mahalanobis distance was propose to estimate missing data and singular data in flight data in allusion to a lack of a imputation method based on Euclid distance. The nearest neighbors were chosen by the Mahalanobis distance between flight data and then entropy was utilized to obtain estimations of missing values. The estimated values were used for the later estimations. Experiments prove that the method is valid and its performation is higher than the k-nearest neighbors imputation method based on Euclid distance for flight data.
出处
《火力与指挥控制》
CSCD
北大核心
2009年第8期113-115,共3页
Fire Control & Command Control
关键词
飞行数据
数据估计
马氏距离
信息熵
flight data, data estimation, Mahalanobis distance, entropy