ML Models for Energy Efficiency in Office Buildings: A Comprehensive Comparative Analysis

The impact on the environment and rise in energy demand causes significance in achieving energy efficiency. Personalized automation in smart buildings is one of the approaches for efficient management of energy. A data-driven model is used to predict user preferences for electrical appliances. In this work, we compare six different machine learning models to learn user behavior for automating appliances in smart buildings. Pre-processed data sets are used to train and test the model for validating the performance. In our analysis, we found that the Logistic regression and Support vector machine has the highest accuracy and F1-score among other models even though all others have good performance.