The Uncertainty Visualization of Bus Running Time Prediction

The uncertainty is constructed from aleatoric uncertainty and epistemic uncertainty.
- Aleatoric uncertainty can be summarized as a class of errors that can not be avoided. For example, the deviation of bus GPS recording equipment.
- Epistemic uncertainty is the uncertainty that exists in our model. The confidence level of bus running time prediction results is discussed to describe the cognitive uncertainty of the prediction model.
We develop the visualization systems to describe the uncertainty of bus running time prediction, i.e., epistemic uncertainty. In addition, neural network model Bayesian Long Short-Term Memory and Fully connected Layers to predict the bus running time.

Visualization Interface