Probabilistic latent semantic indexing
Proceedings of the 22nd annual international ACM SIGIR conference on Research and development in information retrieval
Machine Learning
The Journal of Machine Learning Research
Item-based top-N recommendation algorithms
ACM Transactions on Information Systems (TOIS)
Tracking dynamics of topic trends using a finite mixture model
Proceedings of the tenth ACM SIGKDD international conference on Knowledge discovery and data mining
Discovering evolutionary theme patterns from text: an exploration of temporal text mining
Proceedings of the eleventh ACM SIGKDD international conference on Knowledge discovery in data mining
Pachinko allocation: DAG-structured mixture models of topic correlations
ICML '06 Proceedings of the 23rd international conference on Machine learning
A mixture model for contextual text mining
Proceedings of the 12th ACM SIGKDD international conference on Knowledge discovery and data mining
Topic sentiment mixture: modeling facets and opinions in weblogs
Proceedings of the 16th international conference on World Wide Web
Emerging topic detection using dictionary learning
Proceedings of the 20th ACM international conference on Information and knowledge management
DIGTOBI: a recommendation system for Digg articles using probabilistic modeling
Proceedings of the 22nd international conference on World Wide Web
| Hi-index | 0.00 |
The popular news aggregator called Digg is a social news service that lets people share new articles or blog postings in web pages with other users and vote thumbs up and thumbs down on the shared contents. Digg itself provides only the functionality to search the articles for the topics provided by users using manually tagged keywords. Helping users to find the most interesting Digg articles with the current hot topics will be very useful, but it is not an easy task to classify the articles according to their topics and discover the articles with the hot topics quickly. In this paper, we propose HotDigg, a recommendation system to provide the articles with hot topics in Digg using a novel probabilistic generative model suitable for representing the activities in Digg service. We next propose an EM algorithm to learn the parameters of our probabilistic model. Our performance study with real-life data from Digg confirms the effectiveness of HotDigg by showing that the articles with current hot topics are recommended.