Taxi Demand-Supply Forecasting: Impact of Spatial Partitioning on the Performance of Neural Networks
This work addresses the need for better spatial partitioning techniques to improve prediction accuracy in neural network models for urban mobility forecasting, representing an incremental improvement.
The paper tackled the problem of forecasting taxi demand and supply by comparing variable-sized Voronoi tessellation and fixed-sized Geohash tessellation using LSTM networks on data from Bengaluru and New York, finding that the variable-sized approach yields superior performance.
In this paper, we investigate the significance of choosing an appropriate tessellation strategy for a spatio-temporal taxi demand-supply modeling framework. Our study compares (i) the variable-sized polygon based Voronoi tessellation, and (ii) the fixed-sized grid based Geohash tessellation, using taxi demand-supply GPS data for the cities of Bengaluru, India and New York, USA. Long Short-Term Memory (LSTM) networks are used for modeling and incorporating information from spatial neighbors into the model. We find that the LSTM model based on input features extracted from a variable-sized polygon tessellation yields superior performance over the LSTM model based on fixed-sized grid tessellation. Our study highlights the need to explore multiple spatial partitioning techniques for improving the prediction performance in neural network models.