id	author	title	date	pages	extension	mime	words	sentences	flesch	summary	cache	txt
cord-318437-tzp33iw7	Lovrić, Mario	Understanding the true effects of the COVID-19 lockdown on air pollution by means of machine learning()	2020-11-06		.txt	text/plain	2851	178	58	In this work, a machine learning approach was designed and implemented to analyze local air quality improvements during the COVID-19 lockdown in Graz, Austria. Concentrations of NO(2) (nitrogen dioxide), PM(10) (particulate matter), O(3) (ozone) and O(x) (total oxidant) were selected from five measurement sites in Graz and were set as target variables for random forest regression models to predict their expected values during the city's lockdown period. However, the primary analysis is based on 97 machine learning (ML) models which were used to capture historical relationships between the 98 attributes and compare the predictions to true pollution values after the COVID-19 lockdowns 99 In order to obtain a realistic picture of air quality during the 104 lockdown, we analyzed the long term measurement data from January 2014 to May 2020 from 105 five measurement sites in the Austria city of Graz (Süd (eng. In this work, we have explored the changes in air pollutant concentrations during the COVID-19 403 lockdown for the city of Graz, Austria.	./cache/cord-318437-tzp33iw7.txt	./txt/cord-318437-tzp33iw7.txt