Enhancing Breast Cancer Detection with Machine Learning: A Predictive Modeling Approach

Abstract

Compared to other methods, the one now used to diagnose breast cancer is not as sensitive or specific. The development of increasingly accurate machine learning algorithms for risk assessment, prediction, and treatment planning has made personalized breast cancer data a reality. The effectiveness of data produced by machine learning algorithms in identifying and categorizing breast cancer is examined in this research. In this post, we will go over a machine-learning strategy that may improve breast cancer diagnosis. To improve the speed and accuracy of diagnoses, our approach utilizes enhanced feature selection methods, reliable classification algorithms, and top-notch model training. After the models were built, we put in a lot of time and effort with the hyperparameters to evaluate various ML approaches. A ROC score of 1.00 for Naive Bayes and a score of 98.10% for Random Forest were the two best models. This study proves that ML algorithms, including the Naive Bayes and random forest methods, can accurately forecast breast cancer outcomes. Machine learning might be used to assess the situation in the future.