Robust Classification of Breast Cancer with Support Vector Machine

Among women worldwide, breast cancer is among the most often occurring and deadly diseases. Improving survival rates depends mostly on early diagnosis, which has driven the application of Machine Learning(ML) methods for automated breast cancer case classification. This work uses the Wisconsin Breast Cancer Dataset to explore Support Vector Machine (SVM) application for breast cancer prediction. Included into the dataset are thirty features extracted from digital images of fine needle aspirates of breast masses: radius, texture, and perimeter. Preprocessing and feature scaling produced training and test subsets of the dataset. Running with a radial basis function (RBF) kernel and through grid search, the SVM model on the test set had an accuracy of 98.24%. The findings show how well SVM manages high-dimensional, complex data and achieves significant accuracy in binary classification issues. These findings highlight the importance of SVM for medical diagnosis applications, particularly for the prediction of breast cancer.