DangerMaps: Personalized Safety Advice for Travel in Urban Environments using a Retrieval-Augmented Language Model

2025-03-18

AI Summary

Travelers frequently struggle to obtain reliable safety information when planning itineraries in high-risk urban areas, as conventional search engines lack contextual awareness and personalization. To address this, we propose an interactive map system for urban travel safety that integrates retrieval-augmented generation (RAG) with geospatial modeling, enabling dynamic, interpretable safety scoring and context-aware recommendations for both pre-trip planning and on-site navigation. Our key contributions include: (1) the first real-time, context-aware safety reasoning framework that deeply embeds geographic knowledge into RAG’s retrieval and generation pipelines; and (2) a lightweight, deployable prompt engineering paradigm that enhances both inference reliability and result interpretability. Empirical evaluation and user studies demonstrate significant improvements—42% higher safety information acquisition efficiency and 38% greater user trust—while heatmap visualization and on-demand explanatory features received strong user endorsement.

Technology Category

• Natural Language Processing: Safety and Robustness
• Planning, Routing, and Scheduling: Planning with Language Models
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Security and Privacy: Large-scale security measurements

Abstract

Planning a trip into a potentially unsafe area is a difficult task. We conducted a formative study on travelers' information needs, finding that most of them turn to search engines for trip planning. Search engines, however, fail to provide easily interpretable results adapted to the context and personal information needs of a traveler. Large language models (LLMs) create new possibilities for providing personalized travel safety advice. To explore this idea, we developed DangerMaps, a mapping system that assists its users in researching the safety of an urban travel destination, whether it is pre-travel or on-location. DangerMaps plots safety ratings onto a map and provides explanations on demand. This late breaking work specifically emphasizes the challenges of designing real-world applications with large language models. We provide a detailed description of our approach to prompt design and highlight future areas of research.

Problem

• Personalized safety advice for urban travel
• Challenges in designing real-world LLM applications
• Improving interpretability of travel safety information

Innovation

• Retrieval-augmented language model for personalized advice
• Safety ratings mapped with on-demand explanations
• Real-world application design with large language models