EP-MEANS: an efficient nonparametric clustering of empirical probability distributions.
Keith HendersonBrian GallagherTina Eliassi-RadPublished in: SAC (2015)
Keyphrases
- probability distribution
- information theoretic
- clustering algorithm
- data driven
- clustering method
- data clustering
- k means
- unsupervised learning
- document clustering
- random variables
- outlier detection
- graph theoretic
- data mining tasks
- categorical data
- high dimensional data
- theoretical analysis
- mutual information
- bayesian networks
- markov random field
- data points