Cancer data classification using binary bat optimization and extreme learning machine with a novel fitness function.

Area of Science:

• Bioinformatics
• Computational Biology
• Machine Learning in Medicine

Background:

• Cancer classification is vital in medicine.
• High-dimensional gene expression data presents challenges for machine learning models.
• Effective feature selection is crucial for accurate cancer classification.

Purpose of the Study:

• To propose a novel methodology for cancer classification using gene expression data.
• To enhance feature selection for high-dimensional gene expression datasets.
• To improve the performance of machine learning models in cancer identification.

Main Methods:

• Utilizing the binary bat algorithm (BBA) for optimized feature selection.
• Employing the extreme learning machine (ELM) for cancer classification.
• Developing and applying a novel fitness function to optimize the BBA for feature selection.

Main Results:

• The proposed methodology, incorporating a novel fitness function for BBA, demonstrates superior performance compared to the original fitness function.
• The optimized feature selection process significantly enhances cancer classification accuracy.
• Experimental results validate the effectiveness of the combined BBA and ELM approach.

Conclusions:

• The proposed bio-inspired feature selection method combined with ELM offers a powerful approach for cancer classification.
• This methodology effectively addresses the challenges posed by high-dimensional gene expression data.
• The novel fitness function significantly improves the performance of the binary bat algorithm in this application.