Non-negative multiple matrix factorization with social similarity for recommender systems.
Guoying ZhangMin HeHao WuGuanghui CaiJianhong GePublished in: BDCAT (2016)
Keyphrases
- matrix factorization
- recommender systems
- collaborative filtering
- tensor factorization
- low rank
- rating prediction
- data sparsity
- cold start problem
- negative matrix factorization
- stochastic gradient descent
- data matrix
- user preferences
- probabilistic matrix factorization
- factorization methods
- nonnegative matrix factorization
- information overload
- personalized ranking
- item recommendation
- implicit feedback
- user feedback
- social interaction
- social media
- missing data
- binary matrix
- variational bayesian
- user generated content
- cold start
- least squares
- user interests
- similarity measure
- recommendation algorithms
- recommendation systems
- social networks