Hybrid Recommender Systems: Survey and Experiments
User Modeling and User-Adapted Interaction
Eigentaste: A Constant Time Collaborative Filtering Algorithm
Information Retrieval
TOAST: a topic-oriented tag-based recommender system
WISE'11 Proceedings of the 12th international conference on Web information system engineering
Using time topic modeling for semantics-based dynamic research interest finding
Knowledge-Based Systems
Hi-index | 0.00 |
In reality, users rarely dedicate excessive interest into only one topic over a long time. We propose a topic-based hidden Markov model to analyze temporal dynamics of users' preference. Experiments show that given observations of a new entrant, the proposed model is able to recommend a specific user group he/she can be classified into and also can anticipate what topic he/she will be mostly interested in.