A Study to Evaluate Various Machine Learning Approaches for Breast Cancer Prediction and Detection

Breast cancer, prevalent among women, especially those with a genetic predisposition, presents substantial challenges in understanding and mitigating its impacts. Early detection is paramount to prevent the progression and spread of this disease. This study evaluates ten different machine learning algorithms: Logistic regression, Support Vector Machine, Random Forest, Decision Tree, K-Nearest Neighbor, Autoencoder, Hybrid, Gaussian Mixture Model, Naïve Bayes, and Gradient Boosting, to understand their efficacy and adaptability across various datasets. Our exploration provides insights into the proficiency of these algorithms in breast cancer estimation and early prediction. Naive Bayes, Gradient Boosting, and Random Forest emerged as the superior performers, demonstrating accuracies of 97.37%, 80.36%, and 92.31%, respectively, across three different datasets. Each algorithm was evaluated on three different datasets, thereby offering a comprehensive understanding of their applicability in distinct breast cancer detection environments.