Improved recommender systems by denoising ratings in highly sparse datasets through individual rating confidence.
Nima JoorablooMahdi JaliliYongli RenPublished in: Inf. Sci. (2022)
Keyphrases
- recommender systems
- collaborative filtering
- denoising
- netflix prize
- recommendation algorithms
- user ratings
- rating prediction
- user preferences
- collaborative filtering recommender systems
- collaborative filtering algorithms
- cold start problem
- cold start
- personalized recommendation
- matrix factorization
- image denoising
- basis pursuit
- recommendation quality
- active user
- user item rating
- total variation
- item based collaborative filtering
- natural images
- noisy images
- high dimensional
- recommendation systems
- latent factor models
- user model
- prediction accuracy
- data sparsity
- product recommendation
- rating matrix
- image processing
- information overload
- user behavior
- collaborative recommendation
- user profiling
- denoising algorithm
- sentiment analysis
- user profiles
- implicit feedback
- user modeling
- wavelet packet
- user feedback
- computer vision