Unsupervised expectation-maximization algorithm initialization for mixture models: A complex network-driven approach for modeling financial time series.
Carlo MariCristiano BaldassariPublished in: Inf. Sci. (2022)
Keyphrases
- expectation maximization
- mixture model
- em algorithm
- unsupervised learning
- probabilistic model
- gaussian mixture model
- complex networks
- maximum likelihood
- k means
- finite mixture model
- parameter estimation
- learning algorithm
- gaussian mixture
- density estimation
- generative model
- automatic model selection
- probability density function
- training data
- model based clustering
- machine learning
- high dimensional
- image segmentation
- mixture modeling