| Keywords |
Machine learning; Airtightness; Predictive modeling; Regression analysis |
| Abstract |
In this study, as a building energy performance evaluation value correction concept, a building airtightness prediction model was developed using a machine learning algorithm capable of regression analysis based on building airtightness data collected from literature and field measurements. R2, MAE, MSE, and RMSE were used to evaluate the accuracy of the prediction model, and the RReliefF algorithm was used to evaluate the contribution of the building elements used in predicting building airtightness. The random forest model had the highest prediction accuracy. Among the building elements, the wall-to-window ratio had the most effect on the airtightness performance. |