An extensive analysis of the interaction between missing data types, imputation methods, and supervised classifiers
作者：
Highlights：
• Interactions between missing data, imputation method and classifier are investigated.
• Interaction between imputation method complexity and classifier can be deduced.
• Complex imputation provides superior consistent results over simple methods.
• Certain combinations of the three components studied produce distinguished behaviors.
• Different behaviors of the imputers for varying amount of missingness are reported.
摘要
•Interactions between missing data, imputation method and classifier are investigated.•Interaction between imputation method complexity and classifier can be deduced.•Complex imputation provides superior consistent results over simple methods.•Certain combinations of the three components studied produce distinguished behaviors.•Different behaviors of the imputers for varying amount of missingness are reported.
论文关键词：Missing data,Imputation methods,Supervised classifiers,Machine learning
论文评审过程：Received 14 November 2016, Revised 23 June 2017, Accepted 15 July 2017, Available online 17 July 2017, Version of Record 24 July 2017.
论文官网地址：https://doi.org/10.1016/j.eswa.2017.07.026
