NAICS-Aware Graph Neural Networks for Large-Scale POI Co-visitation Prediction: A Multi-Modal Dataset and Methodology

2025-07-25

AI Summary

Traditional geographic distance models fail to capture co-visit relationships between heterogeneous points of interest (POIs)—e.g., cafés and fine-dining restaurants—under extreme data sparsity. Method: We propose an NAICS-aware GraphSAGE model that learns industry-code embeddings to encode fine-grained commercial semantics, jointly incorporating spatial distribution, socioeconomic attributes, and spatiotemporal dynamics; we further design a state decomposition strategy to enable efficient training on billion-scale POI graphs. Contribution/Results: Evaluated on a large-scale POI-Graph dataset comprising 94.9 million records, our model achieves an R² of 0.625 (+0.382 over baselines) and improves NDCG@10 by 32%, significantly outperforming purely spatial or single-semantic approaches. This work is the first to deeply integrate a fine-grained industry classification system (NAICS) into graph neural networks, enabling semantic-spatial co-modeling of co-visit patterns and establishing a novel paradigm for urban planning and location intelligence.

Technology Category

• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Machine Learning: Graph-based Machine Learning

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding

Abstract

Understanding where people go after visiting one business is crucial for urban planning, retail analytics, and location-based services. However, predicting these co-visitation patterns across millions of venues remains challenging due to extreme data sparsity and the complex interplay between spatial proximity and business relationships. Traditional approaches using only geographic distance fail to capture why coffee shops attract different customer flows than fine dining restaurants, even when co-located. We introduce NAICS-aware GraphSAGE, a novel graph neural network that integrates business taxonomy knowledge through learnable embeddings to predict population-scale co-visitation patterns. Our key insight is that business semantics, captured through detailed industry codes, provide crucial signals that pure spatial models cannot explain. The approach scales to massive datasets (4.2 billion potential venue pairs) through efficient state-wise decomposition while combining spatial, temporal, and socioeconomic features in an end-to-end framework. Evaluated on our POI-Graph dataset comprising 94.9 million co-visitation records across 92,486 brands and 48 US states, our method achieves significant improvements over state-of-the-art baselines: the R-squared value increases from 0.243 to 0.625 (a 157 percent improvement), with strong gains in ranking quality (32 percent improvement in NDCG at 10).

Problem

• Predicting co-visitation patterns across millions of venues
• Addressing data sparsity and complex spatial-business relationships
• Integrating business taxonomy to improve co-visitation prediction accuracy

Innovation

• NAICS-aware GraphSAGE for co-visitation prediction
• Integrates business taxonomy with learnable embeddings
• Scales via state-wise decomposition of massive datasets