Improving Maritime Traffic Emission Estimations on Missing Data with Conditional RBMs

"Improving Maritime Traffic Emission Estimations on Missing Data with Conditional RBMs" paper code and data for reproducibility.

If you use any of this work, please consider citing or mentioning our work.

Repository organization

The repository is organized as follows:

• DatasetGeneration: script to generate the activations and clustering datasets
• Experiments
  • ARIMA: Code to train the multivariate ARIMAs
  • NavStatus: Code and notes for checking the navigational status correlations
  • ShipTypePrediction: Code used to predict the ship type
  • MainEnginePrediction: Code used to predict the main engine power
• Emission modeling: scripts to generate the final emission result

Contact: alberto.gutierrez(at)bsc.es