An Experimental Study on Decomposition-Based Deep Ensemble Learning for Traffic Flow Forecasting

TL;DR

This study evaluates decomposition-based deep ensemble learning methods for traffic flow forecasting, demonstrating their superior performance over non-decomposition approaches across multiple datasets, while highlighting their sensitivity to aggregation and forecast horizon.

Contribution

It provides a comparative analysis of decomposition-based versus non-decomposition-based deep ensemble methods for traffic prediction, filling a gap in existing research.

Findings

Decomposition-based methods outperform non-decomposition ones in accuracy.

Performance is sensitive to aggregation strategies.

Effectiveness varies with forecasting horizon.

Abstract

Traffic flow forecasting is a crucial task in intelligent transport systems. Deep learning offers an effective solution, capturing complex patterns in time-series traffic flow data to enable the accurate prediction. However, deep learning models are prone to overfitting the intricate details of flow data, leading to poor generalisation. Recent studies suggest that decomposition-based deep ensemble learning methods may address this issue by breaking down a time series into multiple simpler signals, upon which deep learning models are built and ensembled to generate the final prediction. However, few studies have compared the performance of decomposition-based ensemble methods with non-decomposition-based ones which directly utilise raw time-series data. This work compares several decomposition-based and non-decomposition-based deep ensemble learning methods. Experimental results on three traffic datasets demonstrate the superiority of decomposition-based ensemble methods, while also revealing their sensitivity to aggregation strategies and forecasting horizons.