Dynamic Bayesian networks with Gaussian mixture models for short-term passenger flow forecasting.
Jérémy RoosStéphane BonnevayGérald GavinPublished in: ISKE (2017)
Keyphrases
- short term
- dynamic bayesian networks
- gaussian mixture model
- passenger flow
- long term
- mixture model
- bayesian networks
- forecasting model
- load forecasting
- approximate inference
- feature vectors
- em algorithm
- particle filtering
- medium term
- maximum likelihood
- structure learning
- expectation maximization
- feature space
- state variables
- arima model
- graphical models
- machine learning
- image segmentation
- maximum likelihood estimation
- prior knowledge