Topic Discovery through Data Dependent and Random Projections.
Weicong DingMohammad Hossein RohbanPrakash IshwarVenkatesh SaligramaPublished in: ICML (3) (2013)
Keyphrases
- data dependent
- random projections
- topic discovery
- hash functions
- text analysis
- latent dirichlet allocation
- topic models
- dimensionality reduction
- text classification
- random sampling
- dimension reduction
- sparse representation
- principal component analysis
- original data
- similarity search
- low dimensional
- text mining
- document clustering
- image reconstruction
- co occurrence
- pattern recognition
- neural network
- high dimensional data
- generative model
- natural language
- face recognition
- feature selection
- machine learning