Facing the cold start problem in recommender systems

作者：

Highlights：

• We propose a RS that combines classification, semantic and heuristic methods.

• We propose the use of known classification algorithms for creating groups of users.

• Semantic similarity techniques result the similarity between users.

• We propose a weighted scheme for resulting the final prediction of ratings.

摘要

•We propose a RS that combines classification, semantic and heuristic methods.•We propose the use of known classification algorithms for creating groups of users.•Semantic similarity techniques result the similarity between users.•We propose a weighted scheme for resulting the final prediction of ratings.

论文关键词：Recommender systems,Cold start problem

论文评审过程：Available online 16 September 2013.

论文官网地址：https://doi.org/10.1016/j.eswa.2013.09.005