Optimizing Gene Selection and Cancer Classification with Hybrid Sine Cosine and Cuckoo Search Algorithm

Gene expression datasets contain extensive data for exploring various biological processes, yet the presence of redundant and irrelevant genes poses a challenge in identifying crucial ones within high-dimensional biological data. To address this, diverse feature selection (FS) methods have been introduced. Enhancing the efficiency and accuracy of FS techniques is vital for selecting significant genes within intricate multidimensional biological information. In this context, we propose an innovative strategy named the Sine Cosine and Cuckoo Search Algorithm (SCACSA) applicable to popular machine learning classifiers like K-Nearest Neighbors (KNN), Support Vector Machine (SVM), and Naive Bayes (NB). The efficacy of the hybrid gene selection algorithm is assessed using breast cancer dataset and benchmarked against alternative feature selection techniques. Empirical results demonstrate SCACSA superiority in accuracy metrics such as precision, sensitivity, and specificity. Furthermore, the SCACSA approach showcases computational efficiency and consistency, setting it apart from other methods in terms of variability. Given the significance of gene selection in complex biological datasets, SCACSA emerges as a valuable tool for cancer dataset classification, aiding medical professionals in informed decision-making for cancer diagnosis. aiding medical professionals in informed