Cautious relational clustering: A thresholding approach
作者：
Highlights：
• We address the problem of clustering relational data in a cautious way.
• The input of our approach is a relational matrix.
• We first make the matrix self-consistent and then complete it using the data.
• Missing relations are left as such in order to allow for cautious inference.
• Our approach proves to be well-suited to scarce or inconsistent input information.
摘要
•We address the problem of clustering relational data in a cautious way.•The input of our approach is a relational matrix.•We first make the matrix self-consistent and then complete it using the data.•Missing relations are left as such in order to allow for cautious inference.•Our approach proves to be well-suited to scarce or inconsistent input information.
论文关键词：Partial clustering,Relational data,Reliable inference.
论文评审过程：Received 15 January 2019, Revised 22 July 2019, Accepted 23 July 2019, Available online 24 July 2019, Version of Record 1 August 2019.
论文官网地址：https://doi.org/10.1016/j.eswa.2019.112837
