Scaffoldings and Spines: Organizing High-Dimensional Data Using Cover Trees, Local Principal Component Analysis, and Persistent Homology.
Paul BendichEllen GasparovicChristopher J. TralieJohn HarerPublished in: CoRR (2016)
Keyphrases
- high dimensional data
- principal component analysis
- persistent homology
- dimensionality reduction
- low dimensional
- linear discriminant analysis
- lower dimensional
- dimension reduction
- high dimensional
- topological features
- high dimensionality
- principal components
- nearest neighbor
- data sets
- decision trees
- similarity search
- subspace clustering
- manifold learning
- morse theory
- computational geometry
- data points
- data analysis
- high dimensional spaces
- clustering high dimensional data
- locally linear embedding
- feature extraction
- face images
- sparse representation
- feature space
- singular value decomposition
- face recognition
- tree structure
- feature selection
- computer vision
- social networks
- dimensionality reduction methods
- input data
- dimensional data
- multi dimensional
- knn
- mobile robot
- pairwise
- pattern recognition
- neural network