Energy Efficiency in Buildings Using Multivariate Extreme Gradient Boosting
Triando Hamonangan Saragih, Rahmat Ramadhani, Muhammad Itqan Mazdadi, Muhammad Haekal · 2022 Seventh International Conference on Informatics and Computing (ICIC) · 2022
Humans need energy. Energy is often used in our daily activities, from helping with work and household chores to lighting our homes and on the street. Waste of light energy often occurs in households and large buildings. Lots of people do wasteful things like turning on all the lights in the room to make it look bright when they're not needed. A lot of research has recently been done on concerns about wasting energy and its long-term negative effects on the environment. A previous study in 2012 by Tsanas and Xifara used statistical machine learning to assess the energy efficiency of buildings. Their research focuses on calculating results individually rather than calculating all results directly. Other studies were also conducted using Multivariate Random Forest, resulting in a fairly good performance. In this study using the Multivariate Extreme Gradient Boosting method. The concept of this method is the same as the Extreme Gradient Boosting method, but is suitable for multivariate data types. The best results were obtained with MSE 0.537471 and RMSE 0.733124 733124 with gamma parameters 0.1, n-estimator 200 and a combination of training and testing data 90:10 by testing the data without normalization. Based on this test, it proves that Multivariate Extreme Gradient Boosting can produce better performance than previous studies.