LWI-SVD: low-rank, windowed, incremental singular value decompositions on time-evolving data sets.
Xilun ChenK. Selçuk CandanPublished in: KDD (2014)
Keyphrases
- low rank
- singular values
- singular value decomposition
- data sets
- matrix decomposition
- singular vectors
- high dimensional data
- matrix completion
- low rank matrix
- least squares
- kernel matrix
- dimensionality reduction
- rank minimization
- reconstruction error
- low rank approximation
- principal component analysis
- convex optimization
- matrix factorization
- data matrix
- trace norm
- high order
- linear combination
- higher order
- training set
- training data
- missing data
- feature points
- nearest neighbor
- object recognition
- face recognition
- decision trees