Scholarlink

<p>Community detection algorithms have become a valuable tool for analyzing complex social networks. It assists scholars in comprehending network topology intuitively and finding hidden human relations in society network. On the contrary, the community hidden problem is raised because some special organizations need to cooperate in social networks while avoiding discovery by community detection algorithms. To solve this problem, a new safety-based community hiding algorithm is advanced in this paper, which hides the target community by disrupting a limited number of links. First, a safety gain function was designed to find the appropriate link by measuring the denseness of the community. Subsequently, we conducted extensive experiments using eight datasets and five classical community detection algorithms, comparing our method with the latest community hiding algorithms. The experimental results confirm that our proposed algorithm is more efficient than previous community hiding algorithms.</p>

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

Community Hiding Based on Safety Index in Social Networks

  • 期刊论文

  • 发表日期：

    2025年1月17日

查看全文
  • Mengke Huang,
  • Yongbo Li,
  • Yanwei Wang,
  • Dong Liu

摘要

Community detection algorithms have become a valuable tool for analyzing complex social networks. It assists scholars in comprehending network topology intuitively and finding hidden human relations in society network. On the contrary, the community hidden problem is raised because some special organizations need to cooperate in social networks while avoiding discovery by community detection algorithms. To solve this problem, a new safety-based community hiding algorithm is advanced in this paper, which hides the target community by disrupting a limited number of links. First, a safety gain function was designed to find the appropriate link by measuring the denseness of the community. Subsequently, we conducted extensive experiments using eight datasets and five classical community detection algorithms, comparing our method with the latest community hiding algorithms. The experimental results confirm that our proposed algorithm is more efficient than previous community hiding algorithms.