Smart Air-Quality Detection Using Regression Models

As industrialization and urbanization accelerate, the necessity for real-time air quality monitoring intensifies due to increasing air pollution levels. This paper analyzes existing datasets by examining the potential impact of air quality on university professors' performance. The study is a foundational step toward deploying Internet of Things (IoT) devices in professors' office environments. Various Machine Learning Regression Models, including linear regression, ridge regression, random forest, and XGBoost, are evaluated across these datasets. The analyses aim to enhance Air Quality Monitoring Systems (AQMS) by incorporating data processing capabilities, improving their predictive accuracy. Results indicate significant improvements in model performance, with XGBoost demonstrating over 90% accuracy in predicting air quality under different environmental conditions. This research sheds light on how air quality variations affect cognitive and professional performance in academic settings, aiming to guide future interventions to optimize workplace environments for better health and productivity.