Naive Bayes for Smart Building Management: Predicting Workspace Occupancy

¹ I2SP Research Team, Physics Department, Faculty of Sciences Semlalia, Cadi Ayyad University, Marrakesh, Morocco
² Industrial Engineering Department, National School of Applied Sciences, Ibn Zohr University, Agadir, Morocco
³ Univ. Grenoble Alpes, CEA, Liten, Campus Ines, 73375, Le Bourget-du-Lac, France

^* Corresponding author E-mail address: m.ennejjar.ced@uca.ac.ma

Abstract

Occupancy detection plays a crucial role in building management, by improving living conditions and optimizing energy efficiency. So, our paper is a part of this perspective and is divided into two parts. Initially, we delve into the significance of detecting occupancy in buildings, emphasizing its positive impact on human well-being and productivity. Subsequently, the second section is dedicated on using the Naive Bayes Classifier (NBC) to predict occupancy in an office room using variables like temperature, humidity, humidity ratio, light, and CO₂ level. This approach demonstrates an impressive accuracy of 97.7%, underscoring the efficacy and the effectivness of this probabilistic classifier in managing building occupancy.