Interactive and nonparametric modeling of preferences on an ordinal scale using small data

作者：

Highlights：

• MCDA sorting techniques fail to model complex preferences having interactions.

• Interactive nonparametric statistical learning for preference modeling is proposed.

• Algorithmic recommendations are made for different preference structures.

• Modeling is performed incrementally, starting with a small reference set.

摘要

•MCDA sorting techniques fail to model complex preferences having interactions.•Interactive nonparametric statistical learning for preference modeling is proposed.•Algorithmic recommendations are made for different preference structures.•Modeling is performed incrementally, starting with a small reference set.

论文关键词：Preference modeling,Sorting,Active learning,Interactive approach,Multi criteria decision aid

论文评审过程：Received 7 May 2016, Revised 25 August 2016, Accepted 26 August 2016, Available online 29 August 2016, Version of Record 6 September 2016.

论文官网地址：https://doi.org/10.1016/j.eswa.2016.08.063