Assessing Chronic Disease Prediction through Machine Learning Techniques for Breast Cancer Diagnosis: A Comparative Analysis

Abstract

Breast cancer is a lethal illness that affects numerous individuals globally and is constantly evolving. To identify and prevent it early on, a significant amount of research has focused on developing and enhancing data mining techniques. In this context the objective of this study is to assess the efficacy of different machine learning (ML) algorithms in the diagnosis and prediction of breast cancer. Specifically, we will compare the accuracy of techniques such as Deep Learning, Decision Trees, Support Vector Machines, and others. This analysis is particularly important given the substantial impact of breast cancer on women's mortality rates worldwide. The research employs the Wisconsin Breast Cancer Dataset to conduct a comprehensive analysis of various machine learning algorithms for classifying the disease. Performance criteria such as accuracy, sensitivity, and specificity are used to evaluate the diagnostic effectiveness of each model. Ensemble approaches, such as AdaBoost, Random Forest, and Gradient Boost, have proven to be highly accurate, with some configurations obtaining precision rates of up to 100%. This paper stands out by providing a thorough comparative examination of several machine learning algorithms used for breast cancer diagnosis and highlighting the outstanding performance of ensemble approaches. The results underscore the possibility of incorporating these sophisticated machine learning techniques into medical environments, leading to substantial enhancements in the early identification and management approaches for breast cancer. This underscores the significance of ongoing advancements and verification of machine learning models to augment the precision of diagnosis and the quality of care for cancer patients.