A novel graph-based missing values imputation method for industrial lubricant data

Soohwan Jeong, Chonghyo Joo, Jongkoo Lim, Hyungtae Cho, Sungsu Lim, Junghwan Kim

Abstract: Highlights•Missing values are unavoidable in lubricant formulation data in chemical industries.•We propose a novel graph-based imputation method (GBIM).•The method considers the feature information and relations between data points.•The proposed GBIM achieved higher imputation accuracy than the other methods.