Scholarlink

The rapid growth of urban data presents significant opportunities for optimizing resource allocation and economic strategies, yet traditional strategic group analysis overlooks location-specific factors crucial for urban planning. We focus on POI-level strategic group analysis, leveraging semantic and spatial information. However, modeling multi-dimensional strategic relationships is challenging and is further hindered by the limited availability of labeled data. To address these issues, we propose a novel framework, the Hypergraph Conditional Neural Process (HCNP). We represent POI relationships using a heterogeneous urban hypergraph, where strategic group interactions are formulated as hyperedges connecting similar POIs. Additionally, we enhance the meta-learning process with auxiliary tasks to improve performance in sparse data settings. Experiments on three city-wide datasets validate the effectiveness of our approach, offering valuable insights for urban planning and sustainable economic growth.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Optimizing Urban Commercial Landscapes with POI Strategic Group Analysis: A Meta-learning Hypergraph Approach

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Jiaqi Jiang,
  • Xin Lin,
  • Jiahui Jin,
  • Siyuan Huang,
  • Haoyu Chen,
  • Xigang Sun

摘要

The rapid growth of urban data presents significant opportunities for optimizing resource allocation and economic strategies, yet traditional strategic group analysis overlooks location-specific factors crucial for urban planning. We focus on POI-level strategic group analysis, leveraging semantic and spatial information. However, modeling multi-dimensional strategic relationships is challenging and is further hindered by the limited availability of labeled data. To address these issues, we propose a novel framework, the Hypergraph Conditional Neural Process (HCNP). We represent POI relationships using a heterogeneous urban hypergraph, where strategic group interactions are formulated as hyperedges connecting similar POIs. Additionally, we enhance the meta-learning process with auxiliary tasks to improve performance in sparse data settings. Experiments on three city-wide datasets validate the effectiveness of our approach, offering valuable insights for urban planning and sustainable economic growth.