AskNearby: An LLM-Based Application for Neighborhood Information Retrieval and Personalized Cognitive-Map Recommendations

2025-12-02

AI Summary

Existing location-based services operate primarily at the urban scale, neglecting the interplay of spatial, temporal, and cognitive factors in localized decision-making—leading to poor accessibility of nearby places, services, and activities, i.e., the Local Life Information Accessibility (LLIA) problem. To address this, we propose a novel LLIA paradigm and an integrated AI framework unifying proximity-aware retrieval with personalized cognitive map recommendation. Our approach features a three-tier retrieval-augmented generation architecture that jointly leverages graph-structured knowledge, semantic embeddings, and geospatial information for multimodal retrieval, while tightly coupling large language models with user cognitive modeling to enable spatiotemporal–cognitive co-adaptive, context-aware recommendation. Evaluated on real-world community datasets, our method significantly outperforms both general-purpose LLMs and mainstream map-based baselines, achieving substantial gains in retrieval accuracy and recommendation quality—demonstrating superior spatiotemporal localization precision and robustness in cognition-informed ranking.

Technology Category

• Search and Optimization: Local Search
• Machine Learning: Large Multimodal Models (LMMs)
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Search and Retrieval-Augmented AI: Personalized, context-aware and across-device search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

The"15-minute city"envisions neighborhoods where residents can meet daily needs via a short walk or bike ride. Realizing this vision requires not only physical proximity but also efficient and reliable access to information about nearby places, services, and events. Existing location-based systems, however, focus mainly on city-level tasks and neglect the spatial, temporal, and cognitive factors that shape localized decision-making. We conceptualize this gap as the Local Life Information Accessibility (LLIA) problem and introduce AskNearby, an AI-driven community application that unifies retrieval and recommendation within the 15-minute life circle. AskNearby integrates (i) a three-layer Retrieval-Augmented Generation (RAG) pipeline that synergizes graph-based, semantic-vector, and geographic retrieval with (ii) a cognitive-map model that encodes each user's neighborhood familiarity and preferences. Experiments on real-world community datasets demonstrate that AskNearby significantly outperforms LLM-based and map-based baselines in retrieval accuracy and recommendation quality, achieving robust performance in spatiotemporal grounding and cognitive-aware ranking. Real-world deployments further validate its effectiveness. By addressing the LLIA challenge, AskNearby empowers residents to more effectively discover local resources, plan daily activities, and engage in community life.

Problem

• Addresses Local Life Information Accessibility (LLIA) gap in 15-minute cities
• Unifies neighborhood information retrieval with personalized cognitive-map recommendations
• Improves localized decision-making by integrating spatial, temporal, and cognitive factors

Innovation

• RAG pipeline with graph, semantic, and geographic retrieval
• Cognitive-map model encoding user familiarity and preferences
• Unified retrieval and recommendation within 15-minute life circle