NestGNN: A Graph Neural Network Framework Generalizing the Nested Logit Model for Travel Mode Choice

2025-09-08

AI Summary

Traditional nested logit (NL) models suffer from limited representational capacity, struggling to flexibly capture inter-option correlations and non-proportional substitution effects; meanwhile, existing deep learning approaches lack explicit modeling of discrete choice structures. To address this, we propose NestGNN—the first framework integrating graph neural networks (GNNs) into discrete choice analysis. NestGNN constructs an “alternative graph” to explicitly encode dependency relationships among choices and employs a nested utility GNN to jointly learn hierarchical utility functions. While preserving the interpretable two-level nesting structure of NL, NestGNN significantly enhances modeling flexibility and expressive power. Experiments on mode choice prediction demonstrate that NestGNN improves predictive accuracy by 9.2% over baseline NL models. Moreover, it supports elasticity analysis and visualization of substitution patterns, achieving a favorable balance among predictive performance, interpretability, and modeling generality.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Natural Language Processing: Learning & Optimization for NLP
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: Social networks and social learning

Abstract

Nested logit (NL) has been commonly used for discrete choice analysis, including a wide range of applications such as travel mode choice, automobile ownership, or location decisions. However, the classical NL models are restricted by their limited representation capability and handcrafted utility specification. While researchers introduced deep neural networks (DNNs) to tackle such challenges, the existing DNNs cannot explicitly capture inter-alternative correlations in the discrete choice context. To address the challenges, this study proposes a novel concept - alternative graph - to represent the relationships among travel mode alternatives. Using a nested alternative graph, this study further designs a nested-utility graph neural network (NestGNN) as a generalization of the classical NL model in the neural network family. Theoretically, NestGNNs generalize the classical NL models and existing DNNs in terms of model representation, while retaining the crucial two-layer substitution patterns of the NL models: proportional substitution within a nest but non-proportional substitution beyond a nest. Empirically, we find that the NestGNNs significantly outperform the benchmark models, particularly the corresponding NL models by 9.2%. As shown by elasticity tables and substitution visualization, NestGNNs retain the two-layer substitution patterns as the NL model, and yet presents more flexibility in its model design space. Overall, our study demonstrates the power of NestGNN in prediction, interpretation, and its flexibility of generalizing the classical NL model for analyzing travel mode choice.

Problem

• Generalizing nested logit models with graph neural networks
• Capturing inter-alternative correlations in travel choices
• Improving prediction accuracy while retaining substitution patterns

Innovation

• Proposes alternative graph for mode relationships
• Designs nested-utility graph neural network NestGNN
• Generalizes nested logit with neural flexibility