Supervised Low-Rank Semi-nonnegative Matrix Factorization with Frequency Regularization for Forecasting Spatio-temporal Data.
Keunsu KimHanbaek LyuJinsu KimJae-Hun JungPublished in: J. Sci. Comput. (2024)
Keyphrases
- spatio temporal data
- nonnegative matrix factorization
- low rank
- matrix factorization
- matrix decomposition
- trace norm
- semi supervised
- spatio temporal
- sparsity constraints
- negative matrix factorization
- collaborative filtering
- data matrix
- low rank approximation
- missing data
- unsupervised feature selection
- low rank matrix
- data representation
- kernel matrix
- linear combination
- matrix completion
- supervised learning
- moving objects
- databases
- convex optimization
- recommender systems
- singular value decomposition
- least squares
- spectral clustering
- feature selection
- computer vision
- learning algorithm
- machine learning
- singular values
- unsupervised learning
- principal component analysis
- pairwise
- image sequences
- original data
- high order
- objective function
- image processing