Scholarlink

This work presents a novel framework for co-clustering on directional data using angular depth measures and the Depth Medoid Based Clustering Algorithm (DBMCA). The method applies depth-based clustering simultaneously to the rows and columns of data matrices enables the analysis of relationships within and between dimensions. The data in this paper comes from the Italian Ministry of Health’s annual reports on hospital admissions. The case study focuses on interregional hospital mobility for acute care activities under ordinary hospitalization in 2017. The results reveal significant differences and similarities in regional hospitalization patterns, offering insights into healthcare mobility dynamics.

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

A Novel Framework for Co-clustering on Directional Data: Adapting the Depth-Based Medoids Clustering Algorithm (DBMCA) with an Application

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Marco Cardillo,
  • Giuseppe Gismondi,
  • Alessia D’Ambrosio,
  • Antonio D’Ambrosio,
  • Giuseppe Pandolfo

摘要

This work presents a novel framework for co-clustering on directional data using angular depth measures and the Depth Medoid Based Clustering Algorithm (DBMCA). The method applies depth-based clustering simultaneously to the rows and columns of data matrices enables the analysis of relationships within and between dimensions. The data in this paper comes from the Italian Ministry of Health’s annual reports on hospital admissions. The case study focuses on interregional hospital mobility for acute care activities under ordinary hospitalization in 2017. The results reveal significant differences and similarities in regional hospitalization patterns, offering insights into healthcare mobility dynamics.