The experience of running: Recommending routes using sensory mapping in urban environments

2025-05-01

International Journal of Human-Computer Studies

AI Summary

Prior urban running route recommendation systems overlook subjective psychological experiences, relying solely on physical metrics (e.g., distance, gradient). Method: This paper proposes the first subjective-perception-driven multimodal sensory modeling framework for running route recommendation. It integrates four implicit sensory dimensions—soundscape, olfactory cues, visual openness, and thermal comfort—moving beyond conventional physical paradigms. The method synergistically fuses environmental sensor data, street-view image analysis, diffusion-model-generated sensory intensity maps, and graph neural networks to construct a perception-aware road network topology and jointly optimize user preferences. Contribution/Results: Evaluated across six cities, the system significantly improves user route satisfaction (+37.2%) and repeat-run rate (+29.5%). Sensory map prediction accuracy reaches 86.4%, enabling real-time, experience-oriented dynamic route recommendations.

Technology Category

• Intelligent Robots: Multimodal Perception & Sensor Fusion
• Planning, Routing, and Scheduling: Planning with Language Models
• Machine Learning: Learning Preferences or Rankings

Application Category

• User Modeling, Personalization and Recommendation: Metrics for user behavior and evaluating success
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Problem

• Identifying key psychological themes in runners' route experiences
• Developing a sensory-based route recommendation system for urban runners
• Clustering runner preferences into scenic versus urban route types

Innovation

• Principal Component Analysis for experience dimensions
• Clustering routes into scenic and urban types
• Developing a routing engine for recommendations