Real-Time Indoor Air Quality Monitoring and Purification System Using Sensor Data and Machine Learning Technique

Air pollution is a serious threat to human health and the environment, and effective monitoring systems need to be developed to assess air quality in real time. Recent advancements in air quality monitoring have hold on Internet of Things (loT) and machine learning (ML) technologies to increase data collection and predictive capabilities but there are still several challenges like sensor accuracy, data integration, and model optimization. Sensor accuracy is a main issue as low-cost sensors often suffer from calibration drift and environmental interference, leading to unreliable data. Integrating data from various sensors provide a board view of air quality poses significant challenges due to differences in sensor sensitivities and response times and optimizing machine learning models to handle large mixed data while maintaining high predictive accuracy requires continuous refinement. To address these challenges, a real-time indoor air quality monitoring system is proposed using temperature, humidity, dust, methane, carbon monoxide and air quality sensors. Using the Random Forest Master for predictive analysis, the system provides real-time monitoring and daily forecasts. An air filter is installed and automatically activated based on sensor data to immediately respond to poor air quality. The system includes LED indicators, dust purification and air quality classification for accurate data entry. In addition to the disadvantages of air cleaning, it improves environmental monitoring and helps reduce the impact of indoor air pollution on health and well-being.