Tweet, Tweet, Retweet: Conversational Aspects of Retweeting on Twitter
HICSS '10 Proceedings of the 2010 43rd Hawaii International Conference on System Sciences
Short and tweet: experiments on recommending content from information streams
Proceedings of the SIGCHI Conference on Human Factors in Computing Systems
BPR: Bayesian personalized ranking from implicit feedback
UAI '09 Proceedings of the Twenty-Fifth Conference on Uncertainty in Artificial Intelligence
Comparing twitter and traditional media using topic models
ECIR'11 Proceedings of the 33rd European conference on Advances in information retrieval
Effects of user similarity in social media
Proceedings of the fifth ACM international conference on Web search and data mining
| Hi-index | 0.00 |
Retweeting is the key mechanism of information diffusion on microblogging community. It is very challenging for user to choose the suitable tweets for retweeting, given the diverse and massive messages received and limited time on site. Therefore, it is crucial to design a recommender system that automatically recommends tweets for user to retweet. Recommending tweets for retweeting is different from conventional recommender system due to limited explicit feedback, high proportion of cold-start tweets and short tweet active time. In this paper, we propose a novel retweet recommendation (RTR) framework which leverages the implicit feedback to help user find the potential tweets he may want to retweet. RTR is divided into offline learning and online recommendation so that tweets can be taken into account as soon as it is published. In offline learning, we adapt a matrix factorization method based on BPR-OPT framework with implicit feedback to compensate the limited explicit feedback. RTR is able to recommend cold-start tweet based on its content. Extensive experiments on real-world microblogging community clearly show that RTR outperforms upon existing methods.