A novel characterisation-based algorithm to discover new knowledge from classification datasets without use of support
作者：
Highlights：
• We define a catalogue as a classification dataset without duplicate instances.
• Every classification dataset has inside a collection of robust catalogues.
• Both datasets and their derived catalogues contain the same association rules.
• Robust catalogues, i.e. catalogues without uncertainty, have interesting properties.
• Our algorithm gets efficiently all robust catalogues inside a classification dataset.
摘要
•We define a catalogue as a classification dataset without duplicate instances.•Every classification dataset has inside a collection of robust catalogues.•Both datasets and their derived catalogues contain the same association rules.•Robust catalogues, i.e. catalogues without uncertainty, have interesting properties.•Our algorithm gets efficiently all robust catalogues inside a classification dataset.
论文关键词：Classification dataset,Catalogue,Association rules mining,Classification association rules mining
论文评审过程：Received 3 April 2017, Revised 11 October 2017, Accepted 11 October 2017, Available online 13 October 2017, Version of Record 20 October 2017.
论文官网地址：https://doi.org/10.1016/j.eswa.2017.10.029
