Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

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

Luyao Niu, Zhicheng Deng, Boyang Li, Nuoxian Huang, Ruiqi Liu, Wenjia Zhang

TL;DR

The paper defines Local Life Information Accessibility (LLIA) and presents AskNearby, an AI-driven platform that unifies retrieval and personalized recommendations for neighborhood-scale information using a three-layer RAG pipeline (GeoRAG, GraphRAG, VectorRAG) combined with a cognitive map. It formalizes LLIA, details the system architecture, and demonstrates superior retrieval accuracy, spatiotemporal grounding, and reduced hallucinations on real-world Shenzhen data, with supportive field deployments. The contributions advance the realization of the 15-minute city by turning static places into cognitively navigable local ecosystems and enabling timely, context-aware discovery of local resources. The work highlights practical impact for residents and urban communities, while outlining avenues for richer cognitive modeling and broader deployments.

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.

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

TL;DR

The paper defines Local Life Information Accessibility (LLIA) and presents AskNearby, an AI-driven platform that unifies retrieval and personalized recommendations for neighborhood-scale information using a three-layer RAG pipeline (GeoRAG, GraphRAG, VectorRAG) combined with a cognitive map. It formalizes LLIA, details the system architecture, and demonstrates superior retrieval accuracy, spatiotemporal grounding, and reduced hallucinations on real-world Shenzhen data, with supportive field deployments. The contributions advance the realization of the 15-minute city by turning static places into cognitively navigable local ecosystems and enabling timely, context-aware discovery of local resources. The work highlights practical impact for residents and urban communities, while outlining avenues for richer cognitive modeling and broader deployments.

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.

Paper Structure

This paper contains 22 sections, 7 equations, 5 figures, 5 tables.

Table of Contents

  1. Introduction
  2. Related Works
  3. Local Information Retrieval and Recommendation
  4. LLMs for Spatiotemporal Information Retrieval and Recommendation
  5. Methodology
  6. Problem Definition: LLIA
  7. System Overview
  8. RAGs for Local Information Access
  9. Cognitive Map Model for Personalized Recommendation
  10. Experiments and Results
  11. Data Description
  12. Experimental Setup
  13. Evaluation Metrics
  14. Ablation Study
  15. Comparison Experiment
  16. ...and 7 more sections

Figures (5)

  • Figure 1: System overview of AskNearby, combining LLM-based object extraction, multi-source retrieval (geospatial, vector, graph), and a cognitive map model to support active information retrieval and passive local recommendations.
  • Figure 2: Illustration of the recommendation framework in AskNearby, combining semantic relevance, spatial proximity, and public familiarity to compute personalized scores.
  • Figure 3: Spatial distribution of RedNote posts in Shenzhen.
  • Figure 4: Answers to a local life problem from AskNearby and Baidu maps (powered by DeepSeek-R1).
  • Figure 5: Screenshots of the AskNearby system: (a) Recommendation page and (b) Retrieval page.