DISGD: A Distributed Shared-nothing Matrix Factorization for Large Scale Online Recommender Systems.
Heidy HazemAhmed AwadAhmed HassanSherif SakrPublished in: EDBT (2020)
Keyphrases
- matrix factorization
- recommender systems
- collaborative filtering
- low rank
- cold start problem
- data sparsity
- online learning
- factorization methods
- nonnegative matrix factorization
- implicit feedback
- negative matrix factorization
- item recommendation
- missing data
- user preferences
- factor analysis
- personalized ranking
- data matrix
- recommendation quality
- stochastic gradient descent
- tensor factorization
- variational bayesian
- online algorithms
- binary matrix
- recommendation algorithms
- recommendation systems
- probabilistic matrix factorization