Scholarlink

<p>The growth of Internet services and users has made cyberspace an indispensable aspect of human life, utilized to address a vast array of human issues. Opinion leaders (OLs) are Internet users, social network users, and web users with a broad range of activities and a significant impact. OL's influence on other users and the spread of their power have been hotly debated in recent years. Various techniques have been employed to identify these users and exploit their influence in multiple industries. This paper presents an approach to select OLs based on different topological network factors and examines user relationships, trust, and hybridization of the improved grey wolf optimizer (IGWO) and amended salp swarm optimization (ASSO). Additionally, local OLs are chosen using a novel hybrid clustering algorithm. According to formal social network analysis (SNA) parameters, the suggested algorithm yielded (83% and 73% F1, 99% and 98% precision, 72% and 58% recall, and 89% and 75% accuracy) scores for the Bitcoin Alpha and Bitcoin OTC datasets, respectively. The social network marketing (SNM) analysis results for the proposed algorithm on Bitcoin Alpha and Bitcoin OTC datasets are 19% and 32%, respectively, better than other 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

An opinion leader selection in social networks using hybrid amended salp swarm and improved grey wolf optimizer algorithms

  • 期刊论文

  • 发表日期：

    2025年4月22日

查看全文
  • Samad Mohammad Aghdam,
  • Farhad Soleimanian Gharehchopogh,
  • Mohammad Masdari

摘要

The growth of Internet services and users has made cyberspace an indispensable aspect of human life, utilized to address a vast array of human issues. Opinion leaders (OLs) are Internet users, social network users, and web users with a broad range of activities and a significant impact. OL's influence on other users and the spread of their power have been hotly debated in recent years. Various techniques have been employed to identify these users and exploit their influence in multiple industries. This paper presents an approach to select OLs based on different topological network factors and examines user relationships, trust, and hybridization of the improved grey wolf optimizer (IGWO) and amended salp swarm optimization (ASSO). Additionally, local OLs are chosen using a novel hybrid clustering algorithm. According to formal social network analysis (SNA) parameters, the suggested algorithm yielded (83% and 73% F1, 99% and 98% precision, 72% and 58% recall, and 89% and 75% accuracy) scores for the Bitcoin Alpha and Bitcoin OTC datasets, respectively. The social network marketing (SNM) analysis results for the proposed algorithm on Bitcoin Alpha and Bitcoin OTC datasets are 19% and 32%, respectively, better than other algorithms.