Regression clustering with lower error VIA EM algorithm.
Metin VuralPamir ErdemOnur AginPublished in: SIU (2014)
Keyphrases
- em algorithm
- expectation maximization
- model based clustering
- mixture model
- finite mixture models
- mixture modeling
- maximum likelihood
- k means
- matrix factorisation
- generative model
- parameter estimation
- gaussian mixture model
- maximum likelihood estimation
- clustering method
- expectation maximisation
- incomplete data
- hyperparameters
- clustering algorithm
- unsupervised learning
- deterministic annealing
- likelihood function
- regression model
- density estimation
- minimum message length
- probability density function
- gaussian mixture
- information theoretic
- maximum a posteriori
- log likelihood
- penalized likelihood
- model selection
- bayesian model selection
- finite mixture model
- bayesian information criterion
- data clustering
- hierarchical clustering
- mixture distribution
- categorical data
- data points
- self organizing maps
- hidden variables
- linear regression
- log likelihood function
- probabilistic model
- bayesian framework