Uncertainty Quantification for Traffic Forecasting Using Deep-Ensemble-Based Spatiotemporal Graph Neural Networks.
Tanwi MallickJane MacFarlanePrasanna BalaprakashPublished in: IEEE Trans. Intell. Transp. Syst. (2024)
Keyphrases
- neural network
- neural network ensemble
- spatiotemporal databases
- random walk
- backpropagation neural networks
- network traffic
- graph representation
- graph structure
- akaike information criterion
- real time
- competitive learning
- graph theory
- short term
- spatial and temporal
- pattern recognition
- space time
- ensemble methods
- prediction model
- recurrent neural networks
- road network
- traffic volume
- directed graph
- fuzzy logic
- back propagation
- spatiotemporal data
- artificial neural networks
- feature selection
- short term prediction
- graph model
- genetic algorithm
- directed acyclic graph
- long term
- multilayer perceptron
- self organizing maps
- weighted graph
- ensemble learning
- training process
- bipartite graph
- graph databases
- graph matching
- conditional probabilities
- training data
- neural network model
- learning algorithm
- fault diagnosis
- structured data