Analyzing and improving stability of matrix factorization for recommender systems.
Edoardo D'AmicoGiovanni GabboliniCesare BernardisPaolo CremonesiPublished in: J. Intell. Inf. Syst. (2022)
Keyphrases
- matrix factorization
- recommender systems
- collaborative filtering
- low rank
- negative matrix factorization
- factorization methods
- data sparsity
- nonnegative matrix factorization
- cold start problem
- probabilistic matrix factorization
- user preferences
- item recommendation
- factor analysis
- missing data
- variational bayesian
- implicit feedback
- data matrix
- user profiles
- recommendation systems
- information overload
- stochastic gradient descent
- recommendation algorithms
- personalized recommendation
- user feedback
- cold start
- latent factor models
- multiscale
- text mining
- personalized ranking