Machine Learning Classification Algorithms for Accurate Breast Cancer Diagnosis

Abstract

The introduction of algorithms based on machine learning (ML) has revolutionized computer science by allowing machines to learn without explicit programming. These algorithms automate operations, like categorization and clustering, that formerly required human interaction. In the medical field, classification algorithms based on machine learning have been used extensively to diagnose various diseases, including breast cancer, which remains one of the leading causes of death among women. This study describes an experiment that analyzes the effectiveness of machine learning-based algorithms for breast cancer diagnosis. Among the techniques evaluated are Logistic Regression, Decision Tree, Neural Network, Naive Bayes, The kNN, support vector machines (SVM), AdaBoost, Stochastic gradient descent (SGD), CN2 rule inducer, Constant, and Random Forest. The investigation used the renowned Wisconsin Diagnosis Breast Cancer dataset (WDBC). The primary objective of this experiment is to evaluate the Precision and accuracy of these algorithms in discriminating between malignant and benign tumors. The SVM model has the best accuracy at 0.975%, suggesting that Random Forest is a viable breast cancer diagnostic tool. These findings indicate that algorithms based on machine learning can potentially increase the accuracy and speed of breast cancer diagnosis. This research can inform the development of more precise and influential breast cancer diagnostic tools, enabling physicians to administer quick and appropriate therapy to patients.