FROG: Inference from knowledge base for missing value imputation

作者：

Highlights：

• A data imputation framework suitable for values with complex semantics in various types is proposed.

• An index-based knowledge base intrieval is proposed which can speed up computation of the framework.

• An inference algorithm on knowledge base for missing value imputation is designed based on the framework.

• Extensive experiments are performed to show that the proposed methods are effective and efficient.

摘要

•A data imputation framework suitable for values with complex semantics in various types is proposed.•An index-based knowledge base intrieval is proposed which can speed up computation of the framework.•An inference algorithm on knowledge base for missing value imputation is designed based on the framework.•Extensive experiments are performed to show that the proposed methods are effective and efficient.

论文关键词：Knowledge base,Missing values,Inference,Imputation,Data quality

论文评审过程：Received 1 March 2017, Revised 31 December 2017, Accepted 2 January 2018, Available online 4 January 2018, Version of Record 20 February 2018.

论文官网地址：https://doi.org/10.1016/j.knosys.2018.01.005