Relevance-based language modelling for recommender systems
作者：
Highlights：
• A new recommendation approach based on the Relevance Modelling (RM) of the problem is proposed.
• The neighbour selection problem is improved by Posterior Probabilistic Clustering (PPC).
• Background information such as item popularity is successfully integrated by using RM models.
• Performance of the recommendation improves when more clusters are considered in the PPC technique.
• Combination of both contributions leads to an even better performance than their separate application.
摘要
•A new recommendation approach based on the Relevance Modelling (RM) of the problem is proposed.•The neighbour selection problem is improved by Posterior Probabilistic Clustering (PPC).•Background information such as item popularity is successfully integrated by using RM models.•Performance of the recommendation improves when more clusters are considered in the PPC technique.•Combination of both contributions leads to an even better performance than their separate application.
论文关键词：Relevance models,Recommender systems,Collaborative filtering,Probabilistic clustering
论文评审过程：Received 31 July 2012, Revised 5 March 2013, Accepted 5 March 2013, Available online 9 April 2013.
论文官网地址：https://doi.org/10.1016/j.ipm.2013.03.001
