Scholarlink

Particle Swarm Optimization (PSO) is widely acknowledged as one of the most effective swarm intelligence approaches in the field of metaheuristics. This paper introduces an adaptive variant of PSO that leverages fitness landscape information, specifically computing the ruggedness factor. The proposed method aims to identify the optimal PSO strategy by adopting an adaptive rule to update PSO parameters based on the ruggedness factor. The effectiveness of this approach is demonstrated through its evaluation on the feature selection problem, showcasing promising outcomes.

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 Landscape-Driven Particle Swarm Optimization: A Preliminary Study on Feature Selection

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Mohammed El Amrani,
  • Malek Sarhani,
  • Abtin Nourmohammadzadeh,
  • Jawad Abrache

摘要

Particle Swarm Optimization (PSO) is widely acknowledged as one of the most effective swarm intelligence approaches in the field of metaheuristics. This paper introduces an adaptive variant of PSO that leverages fitness landscape information, specifically computing the ruggedness factor. The proposed method aims to identify the optimal PSO strategy by adopting an adaptive rule to update PSO parameters based on the ruggedness factor. The effectiveness of this approach is demonstrated through its evaluation on the feature selection problem, showcasing promising outcomes.