ABSTRACT With the rapid development of smart grid technologies and the Internet of Things (IoT), smart homes have become an important component of modern energy management systems. Efficient prediction of household energy consumption can help optimize energy usage, reduce electricity costs, and improve grid reliability. However, residential energy consumption patterns are highly dynamic and influenced by multiple factors such as weather conditions, occupancy patterns, appliance usage, and time of day. Traditional statistical forecasting techniques often fail to accurately capture these nonlinear relationships. Machine learning algorithms have recently emerged as effective tools for energy consumption prediction. This study proposes a machine learning-based energy consumption forecasting model using the Extreme Gradient Boosting (XGBoost) algorithm. Historical household electricity consumption data along with environmental and temporal variables are used as input features for the prediction model. The performance of the proposed approach is evaluated using Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE). The results indicate that the XGBoost model provides accurate energy consumption predictions and can support intelligent energy management in smart homes. Key words: Smart Home, Energy Consumption Prediction, XGBoost, Machine Learning, Smart Grid.