Drilling fluid is a crucial factor in the drilling operation. The loss of drilling fluid during drilling is inevitable. Attempts to compute based on physical modeling are not feasible, and experimental correlations are of-ten imprecise in calculating the amount of drilling fluid lost. Currently, machine learning is the most effective method to solve the above problem. In this study, the authors will perform the prediction of loss of circulation during drilling using Decision Tree and ensemble learning algorithms including Random Forest, Extra Tree, Gradient Boosting, and XGBoost. The data used in the study were obtained from the Marun oil field in Iran. The prediction results of the algorithms are compared based on statistical parameters such as Mean Absolute Error (MAE), Root Mean Squared Error (RSME), and Coefficient of Determination (R2), and it is found that the Extra Tree algorithm has the most accurate prediction results with the lowest MAE and RMSE (3.6284 and 10.2039), and the highest R2 (0.9891) in the five algorithms mentioned.