A Multi-Armed Bandit Model Selection for Cold-Start User Recommendation.
Crícia Z. FelícioKlérisson V. R. PaixãoCélia A. Zorzo BarcelosPhilippe PreuxPublished in: UMAP (2017)
Keyphrases
- model selection
- cold start
- recommender systems
- cold start problem
- collaborative filtering
- cross validation
- user preferences
- data sparsity
- hyperparameters
- tag recommendation
- parameter estimation
- user ratings
- implicit feedback
- recommendation algorithms
- personalized recommendation
- generalization error
- mixture model
- multi armed bandit
- machine learning
- feature selection
- recommendation systems
- user interests
- information overload
- matrix factorization
- user model
- prediction accuracy
- user feedback
- web search
- posterior distribution
- gaussian mixture model
- user profiles
- relevance feedback
- support vector
- reinforcement learning