Recommender system based on pairwise association rules

作者：

Highlights：

• Recommender system resistant to the cold-start problem is proposed.

• System builds a model of preferences from transactions performed by a population.

• Evaluated on transactional dataset from a real world dietary intake recall system.

• Applications to recommender and ranking tasks are demonstrated.

摘要

•Recommender system resistant to the cold-start problem is proposed.•System builds a model of preferences from transactions performed by a population.•Evaluated on transactional dataset from a real world dietary intake recall system.•Applications to recommender and ranking tasks are demonstrated.

论文关键词：Association rules,Cold-start problem,Data mining,Ontologies,Recommender systems

论文评审过程：Received 4 April 2018, Revised 9 July 2018, Accepted 10 July 2018, Available online 21 August 2018, Version of Record 2 September 2018.

论文官网地址：https://doi.org/10.1016/j.eswa.2018.07.077