加权壳近邻填充数学模型

Weighted Shell-Neighbor Imputation Models

提出加权壳近邻填充(WSNI)缺失数据数学模型,充分利用壳近邻填充选取近邻数据的特性,侧重于被重复选择的近邻点,有效提高了填充效果.实验结果表明,提出的加权壳近邻填充数学模型比k近邻填充和壳近邻填充的效果好.

Weighted shell neighbor imputation （WSNI） is designed to improve the missing data imputation efficiency by utilizing the features of nearest selection and focusing on the multiple-selection of the nearest neighbors. Experimental results show that the proposed algorithms have better performances than k-nearest neighbor imputation and shell-neighbor imputation method.