Medical prognostication involves predicting disease outcomes, including complications in recurrence. Early Breast cancer recurrence (BCR) is characterized by the return of cancer within 5 years following initial treatment. Factors such as tumor size, lymph node involvement and hormone receptor status significantly influence the prognosis of breast cancer patients. Early detection of recurrence, particularly in asymptomatic stages, can lead to more effective interventions and improved patient outcomes. This study employs machine learning (ML) classifiers to identify and analyze the factors most predictive of early BCR. Using datasets like the Wisconsin Breast Cancer Recurrence Dataset and the METABRIC dataset, we are applying multiple feature selection schemes to isolate key variables. These results will underscore the potential of ML learning in healthcare and the potential of increasing long term survival rates. Utilizing the Wisconsin dataset, we find that Support Vector Machine has the highest predictive power in our algorithm selection for general Breast Cancer Recurrence. We begin and will continue to apply the same approach for the analysis of Early Breast Cancer Recurrence using the METABRIC data.