Analysis of Thermal Comfort in Mediterranean Climate Buildings Using Random Forest

This paper presents the working process for predicting thermal comfort using a data-driven model with Random Forest. The proposed model is tested considering the ASHRAE Thermal Comfort Database II, on which the results are based. Such a database comprises thermal comfort information worldwide and is developed to generate comfort prediction models based on additional new variables. The results of this study indicate that this approach has the potential to provide more accurate comfort predictions leading to more efficient and comfortable buildings.