In this paper, we present the application of bagging based ensemble technique i.e., Random Forest to predict survival status of lung cancer patients who underwent major lung resections for primary lung cancer using thoracic surgery. Further we interpret the random forest ensemble machine learning model to find out the reasons for particular prediction also known as explanatory artificial intelligence (XAI).The data used for this paper is thoracic surgery patient’s data, in which data was collected retrospectively at Wroclaw Thoracic Surgery Centre in the year 2007 and 2011. As it is highly imbalanced data so to prevent machine learning classifier to predict biased decision in favor of majority class we used the synthetic minority oversampling technique (SMOTE) to deal with it. The proposed solution combines the benefits of using bagging based ensemble method with identification and interpretation of risk factors in classifying survival status of lung cancer patients which can help assist the medical practitioner in their diagnosis