A Machine Learning Framework for Effective Prediction of Breast Cancer

Breast cancer, a form of cancer originating in breast cells, typically necessitates treatment involving X-ray, chemotherapy, or a blend of these methods. Early detection significantly enhances survival rates, making accurate prediction crucial. Artificial intelligence stands at the forefront of this effort, tackling the formidable task of breast cancer prediction, a pivotal area for clinicians and researchers alike. Using machine learning (ML) models such as Decision tree, Naïve Bayes, and Random Forest 32 variables and 569 observations make up the Wisconsin Breast Cancer Data's breast cancer diagnostic medical dataset. Training, testing, and validation procedures were carried out in accordance with the data analysis methodology. The following parameters were used to assess the models' performance: precision, F1 count, sensitivity, specificity, and accuracy of classification. The three trained models may produce the best classification and prediction outcomes, according to the training and results. The Random forest models achieved 99.75% accuracy, outperforming the other models. Ultimately, the findings underscore the models' clear superiority, demonstrating their ability to yield more precise predictions in breast cancer prognosis.