Scholarlink

<p>Social networks play a vital role in transforming how information is produced, consumed, and shared, making social network analysis a multidisciplinary research focus. Key areas include community detection, online service discovery, and identifying influential nodes. Since no single meta-heuristic algorithm can solve all problems efficiently (as per the No Free Lunch Theory), new algorithms are continually developed to enhance diversity. This study presents the Invasive Weed Optimization Gravitational Search Algorithm (IWOGSA), a hybrid approach that merges the adaptability of Weed Optimization with the precision of Gravitational Search. Designed for continuous optimization, IWOGSA leverages two reproduction mechanisms normal distribution and gravitational dynamics—to balance exploration and exploitation effectively. Building on this, we introduce the Dynamic Invasive Weed Optimization Gravitational Search Algorithm (DIWOGSA), specifically developed for influence maximization in complex networks. DIWOGSA integrates a topology-aware initialization strategy and a local search operator, enabling efficient convergence and enhanced influence spread. Experimental results validate IWOGSA's superiority in continuous optimization benchmarks and demonstrate DIWOGSA's competitive edge in influence maximization tasks, surpassing traditional algorithms in performance and computational efficiency. These findings highlight the potential of IWOGSA and DIWOGSA as robust frameworks for continuous and discrete optimization problems.</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

Maximizing penetration in complex networks via the discretization of a new continuous meta-heuristic algorithm

  • 期刊论文

  • 发表日期：

    2025年4月28日

查看全文
  • Junlei Dong,
  • Haichang Jiang,
  • Zaihui Cao

摘要

Social networks play a vital role in transforming how information is produced, consumed, and shared, making social network analysis a multidisciplinary research focus. Key areas include community detection, online service discovery, and identifying influential nodes. Since no single meta-heuristic algorithm can solve all problems efficiently (as per the No Free Lunch Theory), new algorithms are continually developed to enhance diversity. This study presents the Invasive Weed Optimization Gravitational Search Algorithm (IWOGSA), a hybrid approach that merges the adaptability of Weed Optimization with the precision of Gravitational Search. Designed for continuous optimization, IWOGSA leverages two reproduction mechanisms normal distribution and gravitational dynamics—to balance exploration and exploitation effectively. Building on this, we introduce the Dynamic Invasive Weed Optimization Gravitational Search Algorithm (DIWOGSA), specifically developed for influence maximization in complex networks. DIWOGSA integrates a topology-aware initialization strategy and a local search operator, enabling efficient convergence and enhanced influence spread. Experimental results validate IWOGSA's superiority in continuous optimization benchmarks and demonstrate DIWOGSA's competitive edge in influence maximization tasks, surpassing traditional algorithms in performance and computational efficiency. These findings highlight the potential of IWOGSA and DIWOGSA as robust frameworks for continuous and discrete optimization problems.