IRIME: Mitigating exploitation-exploration imbalance in RIME optimization for feature selection.

Area of Science:

• Computational Intelligence
• Optimization Algorithms
• Machine Learning

Background:

• The Rime Optimization Algorithm (RIME) faces challenges including suboptimal convergence and a poor balance between exploration and exploitation.
• These limitations hinder its effectiveness in solving complex optimization problems.

Purpose of the Study:

• To introduce an enhanced Rime Optimization Algorithm (IRIME) that addresses the limitations of the original RIME.
• To evaluate IRIME's performance on benchmark problems and its applicability to engineering and feature selection tasks.

Main Methods:

• IRIME integrates Soft Besiege (SB), Composite Mutation Strategy (CMS), and Restart Strategy (RS).
• Performance was validated using IEEE CEC 2017 benchmark tests and four engineering problems.
• A binary version, bIRIME, was developed for feature selection.

Main Results:

• IRIME demonstrated superior performance compared to other advanced algorithms on benchmark tests.
• IRIME effectively solved practical engineering problems.
• bIRIME achieved excellent results on diverse datasets for feature selection, outperforming existing methods in subset selection and classification accuracy.

Conclusions:

• IRIME offers significant improvements over the standard RIME algorithm.
• bIRIME shows strong potential for effective feature selection in machine learning applications.