ISTS: Implicit social trust and sentiment based approach to recommender systems
作者：
Highlights：
• Using Online Social Networks as new source of data in products recommendations.
• Modelling implicit trust between friends using microblogging features in Twitter.
• A probabilistic method to compute sentiment in micro-reviews into numerical rating.
• Comparing different regression algorithms to predict ratings for cold-start problem.
• Support Vector Regression shows the highest accuracy in ratings prediction.
摘要
•Using Online Social Networks as new source of data in products recommendations.•Modelling implicit trust between friends using microblogging features in Twitter.•A probabilistic method to compute sentiment in micro-reviews into numerical rating.•Comparing different regression algorithms to predict ratings for cold-start problem.•Support Vector Regression shows the highest accuracy in ratings prediction.
论文关键词：Recommender systems,Machine learning,Trust,Sentiment analysis,Microblogging
论文评审过程：Received 30 March 2015, Revised 3 June 2015, Accepted 18 July 2015, Available online 26 July 2015, Version of Record 29 August 2015.
论文官网地址：https://doi.org/10.1016/j.eswa.2015.07.036
