Iranian Journal of

Introduction: Breast cancer is the most common malignancy among women and the second leading cause of cancer mortality. Gene expression analysis using microarray data reveals molecular patterns associated with disease progression, aiding in diagnosis and treatment. However, the high dimensionality of such data poses significant challenges for machine learning methods.
Methods: Breast cancer is the most common malignancy in women and a major cause of cancer mortality. This study introduces a hybrid feature selection method combining filter, wrapper, FWGPC, and deep learning to handle high-dimensional microarray data. The approach enhances classification accuracy and identifies key genes linked to breast cancer.
Results: Experimental results demonstrate that the proposed method achieves classification accuracies of 99.96% and 96.1% on the BC-TCGA and GSE datasets, respectively. It outperforms many other classification approaches in identifying new breast cancer cases.
Conclusion: This study applies the FWGPC algorithm by combining filter and wrapper approaches for feature selection in breast cancer microarray data. The best outcomes compete, and classification accuracy is evaluated through deep learning. The approach enables key gene identification and improves classification performance.