Artificial Intelligence for Sustainable Applications

4Review of Classification and Feature Selection Methods for Genome-Wide Association SNP for Breast Cancer

L.R. Sujithra¹* and A. Kuntha^2†

¹CSE Dr. N. G. P. Institute of Technology, Coimbatore, India

²Coimbatore Institute of Technology, Coimbatore, India

Abstract

Cancer is a complicated disease with many molecular changes driven by hereditary, environmental, and lifestyle factors. Cancer cells develop abnormalities that alter the cells normal development, proliferation, and death cycle. Breast cancer remains the commonest analyzed cancer among women and the main cause of cancer-related deaths. Selection and identification of Single-Nucleotide Polymorphisms (SNPs) remain most significant assessment for Genome-Wide Association Studies (GWAS) related to breast cancer. Nevertheless, at the domain to detecting SNP and classification of healthy-patient, several significant challenges remain. The greatest difficulty is the problem of dimensionality as, the total amount of observations is significantly less than the total amount of SNPs and the healthy-patient data quantity also differ. Because of these difficulties, selecting and classifying features is extremely challenging. Machine Learning (ML) is a revolutionary method that is ideally situated to uncover the unseen biological interactions used for enhanced breast cancer finding and diagnosis. This review goal is to determine the best effective approach of analyzing SNP data by combining several Feature Selection (FS) and classification ...