Predicting Liner Arrival Time Based on Deep Learning

Sea transportation has become the principal mode of transportation. It is of great significance to accurately predict the estimated time of arrival (ETA) of the liner carriage. This paper proposes a model based on deep learning algorithm to deal with liner arrival time prediction in sea transportation. Two data cleaning algorithms and one data enhancement algorithm are presented, with data cleaning effectively cleaning the GPS data generated by the liner and data enhancement increasing the diversity of data samples. A method based on deep learning to predict liner arrival time is provided, using the Factorization Machine (FM) model to generate second-order crossover features, and grouped convolution and attention mechanisms to enhance the representation ability of the model. Experiments show that the method proposed control the prediction error better than traditional machine learning models.