Heroku App Demo very slow unfortunately

Also read more about projet and see PowerPoint presentation here

Urbanization and industrialization may lead to several problems, including logistics, healthcare, and air-quality issues. Nowadays, the common occurrence of smog caused by industrialization increases has brought environmental pollution to its peak. The fine particulate matter (PM2.5) is one of the risky pollutants. According to the World Health Organization (WHO), almost 90% of people breathe polluted air that exceeds its air-quality guidelines [1]. Likewise, a few hours of repeated exposure to PM2.5 can trigger cardiovascular problems [2]-[4]. Poor air- quality not only threats the health and lives of individuals, but also impacts the economy. The Organization for Economic Cooperation and Development (OECD) estimates that air-pollution costs one percent of the world’s domestic product (GDP) [5]. As a result, effective systems to monitor and predict air-pollution are important tools for health reasons, as well as to support governments and decision makers. Nevertheless, the PM2.5 formation process is very complex, which demands meticulous data analysis methods. Machine learning methods may help in dealing with such endeavors, without requiring knowledge about the physical and/or chemical characteristics of atmospheric pollutants. Moreover, they can combine different data types including air pollutants, meteorological, and macroeconomics data to help us to understand their relationships.

This project aims to use machine learning to predict air-pollution levels by combining air-pollutants, meteorological, and macroeconomics data. Different machine learning techniques will be considered, including random forest, neural networks, decision trees, and clustering techniques using the air pollutant emissions data sets.

References [1] Nada Osseiran, Christian Lindmeier: 9 out of 10 people worldwide breathe polluted air, but more countries are taking action; 2018. Online bit.ly/3I70PkF. Last accessed in February 2022. [2] J. A. Ailshire and E. M. Crimmins, “Fine Particulate Matter Air Pollution and Cognitive Function Among Older US Adults,” Am. J. Epidemiol., vol. 180, no. 4, pp. 359–366, Aug. 2014. [3] U. Pöschl, “Atmospheric Aerosols: Composition, Transformation, Climate and Health Effects,” Angew. Chem. Int. Ed., vol. 44, no. 46, pp. 7520–7540, Nov. 2005. [4] Y. Du, X. Xu, M. Chu, Y. Guo, and J. Wang, “Air particulate matter and cardiovascular disease: the epidemiological, biomedical and clinical evidence,” J. Thorac. Dis., vol. 8, no. 1, Jan. 2016. [5] OECD, The Economic Consequences of Outdoor Air Pollution. OECD, 2016.

Data sets:

• eea.europa.eu/themes/air/air-emissions-data

• Expectation maximization (EM)\Inputing missing data https://scikit-learn.org/stable/modules/impute.html

• LIME, SHAP

• causal inference

• bayesian model