User-controlled federated matrix factorization for recommender systems.
Vito Walter AnelliYashar DeldjooTommaso Di NoiaAntonio FerraraFedelucio NarducciPublished in: J. Intell. Inf. Syst. (2022)
Keyphrases
- matrix factorization
- recommender systems
- collaborative filtering
- item recommendation
- user preferences
- low rank
- cold start problem
- information overload
- data sparsity
- implicit feedback
- negative matrix factorization
- factorization methods
- nonnegative matrix factorization
- recommendation quality
- user model
- cold start
- factor analysis
- user profiles
- data matrix
- probabilistic matrix factorization
- latent factors
- recommendation systems
- variational bayesian
- user feedback
- user interests
- user ratings
- recommendation algorithms
- missing data
- stochastic gradient descent
- personalized recommendation
- tensor factorization
- latent factor models
- high dimensional
- document collections
- image classification