作者: Xavier Ferrer , Yoke Yie Chen , Nirmalie Wiratunga , Enric Plaza
DOI: 10.1007/978-3-319-12069-0_7
关键词: Data mining 、 Sentiment analysis 、 Dynamics (music) 、 Ranking (information retrieval) 、 Preference 、 Temporal information 、 Information retrieval 、 Recommender system 、 Feature (machine learning) 、 Product (category theory) 、 Computer science
摘要: Capturing users’ preference that change over time is a great challenge in recommendation systems. What makes product feature interesting now may become the accepted standard future. Social recommender systems harness knowledge from user expertise and interactions to provide have potential capturing such trending information. In this paper, we model our system using sentiment rich generated reviews temporal Specifically integrate these two resources formalise novel aspect-based ranking captures distribution of aspect sentiments so preferences users time. We demonstrate utility proposed by conducting comparative analysis on data extracted Amazon.com Cnet. show considering leads better