Smart Lamp Control Based on User Behavior for Two Lamps Using Classification and Regression Tree (CART) Algorithm

In the current era, Lamps are equipment that most houses have because they play an important role in daily life. Every year the need for housing continues to increase, and that has an impact on the increasing use of electricity, resulting in excessive expenses. One example of excessive electricity use is forgetting to turn off lights and other electronic equipment while leaving the house. This can be addressed by automating the device in the home by making predictions according to room habits or commonly referred to as the Smart Home. The method used for smart lamps is the Classification and Regression Tree (CART) algorithm. The recorded user habit data will be processed into an information dataset and the data will be processed by the system to make predictions later the data will be retrieved by the device so that the device can run according to the habits of the room. The results obtained in the first test were 93% on both lamps, the second test was 88% on both lamps, the third test was 83% on the internal lamp, and 87% on the outside light. From the prediction, results show that the more data, the greater the accuracy result.