A Copula-Based Classification Using Agglomerated Feature Selection Extraction: An Application in Cervical Cancer Diagnostic

Abstract

The use of gene-expression datasets has significantly enhanced our understanding of complex diseases such as cancer. The importance of the relationship between genes in analyzing such datasets has been highlighted, indicating their crucial role in diagnosing the disease accurately. In this study, we investigate the associated copulas between attributes to extract fundamental block-related components. Subsequently, we perform a classification algorithm based on these components to classify a labeled target variable. Specifically, examining the practical implications and effectiveness of our approach in real-world scenarios, we provide a novel illustrative application in cervical cancer classification.