True mean value discovery over multiple data sources with unknown reliability degrees

作者：

Highlights：

• Identifying the pitfalls and challenges related to estimating population means from multiple sources.

• Proposing a data-driven method for estimating the reliability degree of each source.

• Proposing a novel hypothesis testing procedure that can be applied to multi-source observations.

• Demonstrating the advantages of the proposed algorithm.

摘要

•Identifying the pitfalls and challenges related to estimating population means from multiple sources.•Proposing a data-driven method for estimating the reliability degree of each source.•Proposing a novel hypothesis testing procedure that can be applied to multi-source observations.•Demonstrating the advantages of the proposed algorithm.

论文关键词：Multiple data sources,Source reliabilities,True mean value discovery,Confidence interval

论文评审过程：Received 5 January 2021, Revised 6 April 2021, Accepted 8 April 2021, Available online 16 April 2021, Version of Record 19 April 2021.

论文官网地址：https://doi.org/10.1016/j.knosys.2021.107036