By Stefan Angelov

Statistically, nine out of ten people worldwide are exposed to high levels of air pollutants that lead to serious health problems. The simple act of breathing results in early deaths for millions of people and harms billions more. In fact, since the end of 2018, the WHO has dubbed air pollution the “new tobacco”, while the EU is calling it the “biggest environmental risk” to public health.

The goal of this project is to explore the air quality of the Beijing suburbs dataset and build a prototype model.

Using the pollution data in Beijing, I built a multivariate time series forecasting model that predicts PM2.5 (particulate matter smaller than 2.5 micrometers), O3 (Ozone), SO2 (sulfur dioxide), CO (carbon monoxide), NO2(nitrogen dioxide). I implemented Long Short-Term Memory Networks (LSTMs) which is a type of neural network that adds native support for input data composed of sequences of observations.

LinkedIn: [ Ссылка ]
Project Repo: [ Ссылка ]

свернуть