Breast Cancer Diagnosis Comparative Machine Learning Analysis Algorithms

Abstract

Women have an extremely high chance of contracting breast cancer, which is often undiscovered. It is difficult to pinpoint the genesis of this ailment since it depends on several factors. In contrast, determining whether a cancer is benign or malignant requires a significant amount of effort on the part of doctors and medical professionals. Numerous assessments, such as cell size and uniformity, clump thickness, and other factors, are also taken into consideration. Additionally, this has increased the use of machine learning classifiers and their use as research tools. Using other aspects of artificial intelligence, advanced recovery rooms often depend on equipment that is qualified, while claiming that breast cancer is not. The study's objective is to assess the precision, accuracy, f1 score, specificity, recall, and error rate of machine learning tools that may be used to diagnose breast cancer. In terms of overall dataset correctness, SVM and artificial neural networks have a high accuracy of 97.45% with an error rate of 0.0154.