Scholarlink

<p>Link prediction aims to identify missing or future connections between entities of a complex system when the latter is modeled as a network. This research problem has attracted significant attention due to its relevance in numerous fields. In this work, we propose Network Centrality Link Prediction (NCLP), a novel approach that integrates various network centrality metrics in order to predict the likelihood of future connections between entities of a system. Specifically, we consider weighted betweenness, closeness, and Katz centralities, in addition to the Resource Allocation and Adamic–Adar indices. A genetic algorithm is employed to optimize the weights of these metrics, reflecting their contributions to link prediction. The proposed approach is evaluated on several real-world benchmark networks with varying structural characteristics. Experimental results demonstrate that our method achieves superior predictive performance compared to state-of-the-art link prediction techniques. These findings highlight the effectiveness and robustness of integrating centrality-based measures through an optimization-driven approach, emphasizing the potential of hybrid methodologies for improving link prediction accuracy across diverse network domains. Data and code related to the proposed approach can be found at the following GitHub repository: <a href="https://github.com/elissalichaaelkhoury/NCLP-Link-Prediction">https://github.com/elissalichaaelkhoury/NCLP-Link-Prediction</a>.</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

A novel centrality-based approach for link prediction

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月20日

查看全文
  • Elissa Lichaa El Khoury,
  • Danielle Azar,
  • Eileen Marie Hanna

摘要

Link prediction aims to identify missing or future connections between entities of a complex system when the latter is modeled as a network. This research problem has attracted significant attention due to its relevance in numerous fields. In this work, we propose Network Centrality Link Prediction (NCLP), a novel approach that integrates various network centrality metrics in order to predict the likelihood of future connections between entities of a system. Specifically, we consider weighted betweenness, closeness, and Katz centralities, in addition to the Resource Allocation and Adamic–Adar indices. A genetic algorithm is employed to optimize the weights of these metrics, reflecting their contributions to link prediction. The proposed approach is evaluated on several real-world benchmark networks with varying structural characteristics. Experimental results demonstrate that our method achieves superior predictive performance compared to state-of-the-art link prediction techniques. These findings highlight the effectiveness and robustness of integrating centrality-based measures through an optimization-driven approach, emphasizing the potential of hybrid methodologies for improving link prediction accuracy across diverse network domains. Data and code related to the proposed approach can be found at the following GitHub repository: https://github.com/elissalichaaelkhoury/NCLP-Link-Prediction.