Forgetting techniques for stream-based matrix factorization in recommender systems.
Pawel MatuszykJoão VinagreMyra SpiliopoulouAlípio Mário JorgeJoão GamaPublished in: Knowl. Inf. Syst. (2018)
Keyphrases
- matrix factorization
- recommender systems
- collaborative filtering
- data sparsity
- low rank
- cold start problem
- negative matrix factorization
- factorization methods
- nonnegative matrix factorization
- user preferences
- recommendation quality
- factor analysis
- implicit feedback
- information overload
- item recommendation
- variational bayesian
- latent factor models
- tensor factorization
- probabilistic matrix factorization
- personalized recommendation
- user interests
- personalized ranking
- user profiles
- stochastic gradient descent
- missing data
- factorization method
- recommendation algorithms
- user feedback
- rating prediction