Interest Networks (iNETs) for Cities: Cross-Platform Insights and Urban Behavior Explanations

2025-07-07

AI Summary

This study investigates the effectiveness of user interest networks (iNETs) for urban behavioral modeling using multi-source location data (Google Places and Foursquare) and multi-scale spatial partitioning (H3 hexagonal grids). We propose the iNETs framework and the h3-cities analytical toolkit, integrating graph neural networks, behavioral clustering, and eXplainable AI (XAI) to build an interpretable, hierarchical interest recommendation system. Key findings include: (1) geographic proximity and venue similarity are primary drivers of interest distribution; (2) cross-platform behavioral patterns converge at coarse spatial granularities but diverge significantly at fine scales due to platform-specific biases; and (3) the system effectively distinguishes exploratory versus returning users, recommends high-interest urban regions, and generates natural-language explanations. Contributions include an open-source toolchain and an interactive demonstration platform, substantially enhancing comparability, interpretability, and context-awareness in urban behavior modeling.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Application Domains: Humanities & Computational Social Science

Application Category

• User Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalization
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Content-based information diffusion

Abstract

Location-Based Social Networks (LBSNs) provide a rich foundation for modeling urban behavior through iNETs (Interest Networks), which capture how user interests are distributed throughout urban spaces. This study compares iNETs across platforms (Google Places and Foursquare) and spatial granularities, showing that coarser levels reveal more consistent cross-platform patterns, while finer granularities expose subtle, platform-specific behaviors. Our analysis finds that, in general, user interest is primarily shaped by geographic proximity and venue similarity, while socioeconomic and political contexts play a lesser role. Building on these insights, we develop a multi-level, explainable recommendation system that predicts high-interest urban regions for different user types. The model adapts to behavior profiles -- such as explorers, who are driven by proximity, and returners, who prefer familiar venues -- and provides natural-language explanations using explainable AI (XAI) techniques. To support our approach, we introduce h3-cities, a tool for multi-scale spatial analysis, and release a public demo for interactively exploring personalized urban recommendations. Our findings contribute to urban mobility research by providing scalable, context-aware, and interpretable recommendation systems.

Problem

• Compares iNETs across platforms and spatial granularities for urban behavior
• Analyzes geographic proximity and venue similarity influence on user interest
• Develops explainable recommendation system for high-interest urban regions

Innovation

• Multi-level explainable recommendation system
• Adapts to behavior profiles using XAI
• h3-cities tool for multi-scale analysis