TransFlower: An Explainable Transformer-Based Model with Flow-to-Flow Attention for Commuting Flow Prediction

TL;DR

TransFlower is an explainable transformer model with flow-to-flow attention that significantly improves urban commuting flow prediction accuracy, aiding urban planning and policy-making with interpretable insights.

Contribution

It introduces a novel flow-to-flow attention mechanism within a transformer framework, enhancing prediction accuracy and interpretability for commuting flow modeling.

Findings

Outperforms existing models by up to 30.8% in accuracy.

Incorporates an anisotropy-aware relative location encoder.

Provides interpretable insights into mobility dynamics.

Abstract

Understanding the link between urban planning and commuting flows is crucial for guiding urban development and policymaking. This research, bridging computer science and urban studies, addresses the challenge of integrating these fields with their distinct focuses. Traditional urban studies methods, like the gravity and radiation models, often underperform in complex scenarios due to their limited handling of multiple variables and reliance on overly simplistic and unrealistic assumptions, such as spatial isotropy. While deep learning models offer improved accuracy, their black-box nature poses a trade-off between performance and explainability -- both vital for analyzing complex societal phenomena like commuting flows. To address this, we introduce TransFlower, an explainable, transformer-based model employing flow-to-flow attention to predict urban commuting patterns. It features a geospatial encoder with an anisotropy-aware relative location encoder for nuanced flow representation. Following this, the transformer-based flow predictor enhances this by leveraging attention mechanisms to efficiently capture flow interactions. Our model outperforms existing methods by up to 30.8% Common Part of Commuters, offering insights into mobility dynamics crucial for urban planning and policy decisions.