Time series clustering with an EM algorithm for mixtures of linear Gaussian state space models.
Ryohei UmataniTakashi ImaiKaoru KawamotoShutaro KunimasaPublished in: Pattern Recognit. (2023)
Keyphrases
- em algorithm
- expectation maximization
- mixture model
- parameter estimation
- probabilistic model
- state space
- maximum likelihood
- hidden variables
- mixture components
- gaussian mixture model
- linear gaussian
- incomplete data
- gaussian mixture
- generative model
- mixture distribution
- maximum a posteriori
- maximum likelihood estimation
- bayesian framework
- finite mixture model
- hyperparameters
- prior knowledge
- probability density function
- higher order
- dynamical systems
- image segmentation