Thesis Code: 20010

Thesis Type: M.Sc. thesis in Machine Learning, Data Science, Computer Science, Mathematics, or equivalent

Research Area: Data Science for Industrial and Societal Application

Requirements:

  • Knowledge of Python
  • Software development skills
  • Basic concepts on data science, concerning data analysis, processing and machine learning
  • Basic concepts on image processing

 Description:

During recent years, the topic of air quality is increasingly raising the attention of institutions and governments, for both the impact on health and the direct influence on climate change. Activists and environmentalist organizations are complaining more and more heavily, bringing the topic to the attention of most of the population. A significant proportion of Europe’s population lives in areas, especially cities, where exceedances of air quality standards occur ozone, nitrogen dioxide and particulate matter (PM) pollution pose serious health risks. On purpose, municipalities can benefit from constant monitoring of the pollutants in sensitive areas, for better handling measures to limit their concentration in the air. Moreover, the chance to exploit meteorological data for an estimate of its trend will improve the potential to take the phenomena under control.

This thesis proposes to build a system for processing of the data acquired from the ESA’s newer mission: Copernicus Sentinel-5P. The fleet of satellites continuously acquires information about several pollutants in the atmosphere, since December 2018. The candidate will build a system for the visualization of the aforementioned data, able to display the most recent acquisitions, given a certain area of interest. Moreover, the system will process the information from different sources, considering different features, such as meteorological data and information acquired from other Copernicus spatial missions (i.e. spectral data, SAR data).

Through the examined sources, the candidate will design and develop at least a machine learning/deep learning model to provide an accurate forecast of the evolution of pollutant concentration. It will benefit from a complete dataset and structures to access and process satellite data.

Contact: send a resume with attached the list of exams to alessandro.farasin@linksfoundation.com specifying the thesis code and title.