Scholarlink

This review examines the use of evolutionary optimization, specifically Particle Swarm Optimization (PSO), with artificial neural networks (ANN) incorporating dropout and k-means clustering. It evaluates hybrid approaches for improving clustering accuracy, ANN performance, and robustness across multiple datasets. Findings show that PSO combined with k-means enhances cluster credibility, dropout improves generalization, and advanced PSO variants accelerate convergence. Challenges include computational complexity and limited empirical analysis. These insights highlight the potential of integrated PSO, dropout, and k-means frameworks for optimized neural network and clustering applications.

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 Investigation on Application of Evolutionary Optimisation Techniques on Different ML Models for Developing Credible Clusters

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Mohona Ghosh,
  • Tina Prabhat,
  • Ajit Kumar Pasayat

摘要

This review examines the use of evolutionary optimization, specifically Particle Swarm Optimization (PSO), with artificial neural networks (ANN) incorporating dropout and k-means clustering. It evaluates hybrid approaches for improving clustering accuracy, ANN performance, and robustness across multiple datasets. Findings show that PSO combined with k-means enhances cluster credibility, dropout improves generalization, and advanced PSO variants accelerate convergence. Challenges include computational complexity and limited empirical analysis. These insights highlight the potential of integrated PSO, dropout, and k-means frameworks for optimized neural network and clustering applications.