作者: Svetlin Bostandjiev , John O'Donovan , Tobias Höllerer
关键词:
摘要: This paper presents an interactive hybrid recommendation system that generates item predictions from multiple social and semantic web resources, such as Wikipedia, Facebook, Twitter. The employs techniques traditional recommender literature, in addition to a novel interface which serves explain the process elicit preferences end user. We present evaluation compares different non-interactive strategies for computing recommendations across diverse APIs. Results of study indicate explanation interaction with visual representation increase user satisfaction relevance predicted content.