Leverage Item Popularity and Recommendation Quality via Cost-Sensitive Factorization Machines.
Chih-Ming ChenHsin-Ping ChenMing-Feng TsaiYi-Hsuan YangPublished in: ICDM Workshops (2014)
Keyphrases
- cost sensitive
- recommendation quality
- matrix factorization
- collaborative filtering
- recommender systems
- multi class
- misclassification costs
- recommendation systems
- cost sensitive classification
- cost sensitive learning
- data sparsity
- user preferences
- cold start
- user ratings
- recommendation algorithms
- naive bayes
- class imbalance
- information overload
- active learning
- class distribution
- support vector machine
- personalized recommendation
- active user
- cost sensitive boosting
- feature extraction
- ranking functions
- k nearest neighbor
- knn
- pairwise
- feature space