Breast Cancer Prediction: A Comparative Study of Support Vector Machine and Logistic Regression

Abstract

One of the most common malignancies among women worldwide is breast cancer and a key factor in raising survival rates is early identification. So, it is important to differentiate between malignant (cancerous) or benign (non-cancerous) tumors. Support Vector Machine (SVM) and Logistic Regression are popular machine learning models that has been widely used for binary classification problems including breast cancer prediction. This study explores the effectiveness of SVM and Logistic Regression in predicting breast cancer and compare their performances. This study uses Python programming to implement SVM and Logistic Regression to classify the Breast Cancer Wisconsin dataset from the UCI machine learning repository. Performance metrices such as recall, F1 score, accuracy, precision, and AUC-ROC have all been used to gauge how well these two algorithms work. Upon comparison, the result showed that SVM model outperformed Logistic Regression model on all the performance metrices.