本篇論文針對灰關聯最近鄰居法填補遺失值提出兩方面的改良。在相似度計算方面，考量各屬性與具遺失值屬性的聯關程度，提出以加權式灰關聯法計算計錄間的總體相似度；在資料比對方面，則運用候選資料集取代完整資料集作為加權式灰關聯法搜尋近似資料之集合以節省資料比對次數。根據實驗一結果顯示，在屬性間權重差異大時，加權式灰關聯法的遺失值填補正確率高於灰關聯最近鄰居法。實驗二的結果則說明了整合自動分群之加權式灰關聯法確實能在誤差率差異不大的情況下，有效地降低其資料比對次數。

This paper provides two improvements on the grey-based nearest neighbors approach for missing value prediction. First, in regard to similarity measurement, the weighted grey relational analysis is proposed to calculate the similarity between records. Second, when comparing times, the candidate set is used to reduce the number of comparisons in the complete data set. In the results from experiment 1, the weighted grey relational analysis outperforms the grey-based nearest neighbors approach in accuracy. The result of experiment 2 reveals that the proposed integrating method supplies missing values efficiently and accuracy is not harmed.