Authors : Kadima Muamba Donatien; Mavula Kikwe Alexis; Kako Gbolo Etienne; Mudingumba Kibadi Louqman; Ilunga Mbuyamba Elisée; Mukeba Kalala Magloire; Masumbuku Kashala Willy

Volume/Issue : Volume 11 - 2026, Issue 2 - February

Google Scholar : https://tinyurl.com/bdhh69nz

Scribd : https://tinyurl.com/3mwuburh

DOI : https://doi.org/10.38124/ijisrt/26feb789

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Abstract : Air pollution is a major public health issue in African cities, where monitoring systems remain limited. This study proposes an intelligent air quality monitoring framework that integrates a normative index (Air Quality Index, AQI) and a data mining approach based on the Apriori algorithm. The system was deployed at the Grand Marché in Kinshasa, an area with high population density and heavy traffic. Low-cost sensors were used to measure concentrations of PM₂.₅, PM₁₀, NO₂, SO₂, CO, and NH₃. The data were stored in a relational database and then analyzed according to the dominant pollutant rule to calculate the AQI. The Apriori algorithm was then applied to extract association rules between pollutants and levels of air quality degradation. The results show a marked dominance of fine particles (PM₂.₅ and PM₁₀) in the degradation of the AQI. The association rule analysis reveals frequent co-occurrences between certain gases and episodes of poor air quality, which are invisible in a strictly normative approach. This research demonstrates the relevance of a hybrid model combining regulatory indices and data mining to improve multi-pollutant interpretation in African urban contexts.

Keywords : Air Quality, Apriori, Data Mining, Apriori Algorithm, PM₂.₅, Environmental Monitoring, Kinshasa

References :

1. Chen, X., Wang, Y., & Wang, S. (2020). Comparison of Low-Cost Air Quality Sensors and Reference Instruments in Urban Monitoring Networks. Atmospheric Environment, 224, 117287.
2. Kumar, P., et al. (2020). The Rise of Low-Cost Sensing for Managing Air Pollution in Cities. Environment International, 134, 105212.
3. Maag, B., Zhou, Z., & Thiele, L. (2020). A Survey on Sensor Calibration in Air Quality Monitoring Deployments. IEEE Internet of Things Journal, 7(11), 10–21.
4. Rai, A. C., Snyder, M. G., & Hall, D. (2020). End-User Perspective of Low-Cost Sensors for Outdoor Air Pollution Monitoring. Atmospheric Measurement Techniques, 13(6), 2935–2959.
5. Spinelle, L., Gerboles, M., & Aleixandre, M. (2021). Field Calibration of Low-Cost Sensors for Air Quality Monitoring: Best Practices and Performance Evaluation. Sensors, 21(7), 2427.
6. Zheng, T., et al. (2021). Real-World Evaluation of Low-Cost Particulate Matter Sensors Based on Collocated Data with Reference Monitors. Science of the Total Environment, 795, 148720.
7. Chakraborty, T., & Chakraborty, A. (2021). Machine Learning and Data Mining Approaches in Air Pollution Analysis: A Review. Environmental Challenges, 100099.
8. Statistical Office of the European Union (Eurostat) (2022). Air Quality in Cities — Key Results from the EU Urban Audit. European Environment Agency Report (use as policy reference).
9. Gulia, S., et al. (2022). Characterization of Urban Air Pollution Using Low-Cost Sensors and Data Analytics Techniques. Environmental Monitoring and Assessment, 194(5), 1–16.
10. Casas, N., Payan, D., & Llobet, E. (2023). Recent Advances in Data Mining Applications for Urban Air Quality Assessment. Environmental Science & Technology, 57(14), 5177–5195.
11. Sahu, S. K., et al. (2023). Association Rule Mining to Detect Air Pollutant Interaction Patterns in Mega-Cities. Environmental Pollution, 317, 120754.
12. Liu, H., & Zhang, Y. (2024). Hybrid AQI–Machine Learning Framework for Multi-Pollutant Air Quality Interpretation in Smart Cities. Journal of Cleaner Production, 389, 136528.