A framework for redescription set construction
作者：
Highlights：
• A framework producing large number of highly accurate redescriptions is proposed.
• User-guided multi-objective optimization used for redescription set construction.
• Conjunctive refinement significantly increases redescription accuracy.
• Using Pessimistic Jaccard index can lead to discarding some valuable redescriptions.
• Proposed approach has advantages over currently available approaches.
摘要
•A framework producing large number of highly accurate redescriptions is proposed.•User-guided multi-objective optimization used for redescription set construction.•Conjunctive refinement significantly increases redescription accuracy.•Using Pessimistic Jaccard index can lead to discarding some valuable redescriptions.•Proposed approach has advantages over currently available approaches.
论文关键词：Knowledge discovery,Redescription mining,Predictive clustering trees,Redescription set construction,Scalarization,Conjunctive refinement,Redescription variability
论文评审过程：Received 25 June 2016, Revised 6 October 2016, Accepted 6 October 2016, Available online 8 October 2016, Version of Record 25 October 2016.
论文官网地址：https://doi.org/10.1016/j.eswa.2016.10.012
