Breast cancer recurrence is a challenging issue that profoundly impacts patient well-being, healthcare systems, and society. Modern sensing technologies provide an unprecedented opportunity to gain valuable insights and uncover patterns related to recurrent events. Despite this potential, few studies have delved into survival analysis of breast cancer recurrences based on the extensive healthcare data available. Leveraging this data is crucial for understanding the factors contributing to breast cancer recurrence. This paper introduces an ensemble method known as random survival forest for analyzing time-to-event patterns of breast cancer recurrences using the Surveillance, Epidemiology, and End Results (SEER) data. Our model characterizes the survival probabilities among patients with and without breast cancer recurrences. Ensemble models are constructed by systematically sampling and bootstrapping big data sources. Our experimental findings reveal that the age at the time of cancer recurrence and the time between recurrences follow approximately Gaussian and exponential distributions with means of 61.35 ± 14.03 and 2.61 years, respectively. Moreover, the results identify significant factors such as age, surgical status, tumor stage, and histological grade that influence the likelihood of breast cancer recurrences. This proposed approach in survival analysis holds great potential for aiding healthcare practitioners in the prognosis, treatment, and decision-making regarding breast cancer recurrences.