Breast cancer is the second most common cancer for women in the United States and, accurate identification of cancerous masses is critical to reduce death rates. This thesis explores several classification models that can accurately identify malignant and benign breast masses using data from the from the UCI Machine Learning Repository. In this paper, logistic regression with PCA, random forests, and Naive Bayes are compared to identify the best model for classification. Identifying a powerful classification model can help health care providers streamline screening and diagnosis processes.