Air pollution has become a critical environmental challenge due to rapid urbanization, industrial growth, and increased vehicular emissions. Accurate prediction and visualization of air quality can support timely decision-making and help mitigate negative health and environmental impacts. This project presents a Machine Learning–based system that analyzes real-time environmental sensor data to predict air pollution levels and visualize trends effectively. The system collects key parameters such as PM2.5, PM10, CO₂, NO₂, temperature, and humidity from IOT-based sensors and preprocesses the data to remove noise and missing values. Various machine learning models— including Linear Regression, Random Forest, and LSTM neural networks trained to forecast Air Quality Index (AQI). The best-performing model is deployed to provide short-term pollution predictions .

In this project, an Air Pollution Prediction and Visualization System using machine learning and environmental sensordata was successfully designed and implemented. The system effectively integrates data collection, preprocessing, feature engineering, machine learning-based prediction, AQI calculation, and visualization into a unified framework. Experimental results demonstrate that machine learning and deep learning models are capable of accurately predicting air pollution levels such as PM2.5 and PM10. Among the implemented models, the LSTM model showed superior performance due to its ability to capture temporal patterns in time-series data. The AQI calculation module successfully translated predicted pollutant concentrations into meaningful air quality categories. The visualization module provided clear and interactive representations of both real-time and predicted air quality data, enabling users to easily understand pollution trends and severity levels. This enhances public awareness and supports informed decision-making for environmental management. Overall, the proposed system proves to be reliable, scalable, and effective for air quality monitoring and prediction. It can serve as a valuable tool for researchers, policymakers, and the general public in addressing air pollution challenges and Promoting healthier living environments.

1. Samad, A., et al., Air Pollution Prediction Using Machine Learning Techniques, Atmospheric Environment, 2023. https://www.sciencedirect.com/science/article/pii/S1352231023004132 2.Tran, H. D., et al., Forecasting Hourly PM2.5 Concentration Using Optimized LSTM Networks, Atmospheric Pollution Research, 2023. https://www.sciencedirect.com/science/article/abs/pii/S13522310230058733. Wang, Z., et al., A Spatiotemporal XGBoost Model for PM2.5 Concentration Prediction, Environmental Modelling & Software, 2023. https://pmc.ncbi.nlm.nih.gov/articles/PMC10696222/ 4. Mohammadi, F., et al., Prediction of Atmospheric PM2.5 Levels Using Machine Learning Methods, Sustainability, 2024. https://www.mdpi.com/2071-1050/16/2/911 5.Dai, H., et al., PM2.5 Concentration Prediction Based on Spatiotemporal Deep Learning Models, Sustainability, 2021. https://www.mdpi.com/2071-1050/13/21/12071 6. Li, Y. Peng, “Air Pollution Forecasting Using LSTM De ep Learning Models,” Journal of Atmospheric Environment, 2021. 7.Chen, T., and Guestrin, C., XGBoost: A Scalable Tree Boosting System, ACM SIGKDD, 2016. https://dl.acm.org/doi/10.1145/2939672.2939785 8.Plantower PMS5003 Sensor Datasheet, Plantower Co., Ltd., 2019. 9.World Health Organization (WHO), Air Quality Guidelines, 2021. https://www.who.int/publications/i/item/WHO-HEP-ECH-EHD-21.01 10.Hochreiter, S., and Schmidhuber, J., “Long Short-Term Memory,” Neural Computation, vol. 9, no. 8, pp. 1735–1780, 1997.

© 2025 International Journal of Engineering and Techniques (IJET).

Digital Object Identifier (DOI)

IJET assigns a unique DOI to every accepted article via Zenodo for persistent linking and citation. Your article’s DOI: https://doi.org/{{doi}}

Open DOI About Zenodo DOI

Close