Hyperparameter Optimization for Neural Network based Taxi Demand Prediction

Being able to accurately predict future taxi demand can beneficial not only for taxi companies but also for passengers and the environment as an intelligent taxi planning system can reduce waiting and idle driving times. This work proposes an extended Long Short-Term Memory (LSTM) neural network structure for predicting future taxi demand. Experiments are performed on taxi data from New York City. The model’s hyperparameters are tuned using a very simple selection method based on predictions for only one location at a time. More complex algorithms, hyperopt and BOHB, are implemented to tune the model’s hyperparameters in a more structured and comprehensive way which reduces the prediction error by 2.2% compared to the simple selection method. The results suggest that there are factors that limit the performance gains of popular hyperparameter optimization techniques but also that a relatively simple model can yield useful predictions and outperform several naive benchmark methods.