Modelling the Daily Concentration of Airborne Particles Using 1D Convolutional Neural Networks

Ivan Gudelj; Mario Lovrić; Emmanuel Karlo Nyarko

doi:10.3390/engproc2024068016

Modelling the Daily Concentration of Airborne Particles Using 1D Convolutional Neural Networks

Ivan Gudelj, Mario Lovrić, Emmanuel Karlo Nyarko

Publikation: Bidrag til tidsskrift › Tidsskriftartikel › Forskning › peer review

2 Citationer (Scopus)

Abstract

This paper focuses on improving the prediction of the daily concentration of the pollutants, PM10 and nitrogen oxides (NO, NO2) in the air at urban monitoring sites using 1D convolutional neural networks (CNN). The results show that the 1D CNN model outperforms the other machine learning models (LSTM and Random Forest) in terms of the coefficients of determination and absolute errors.

Originalsprog	Engelsk
Tidsskrift	Engineering Proceedings
Vol/bind	68
Udgave nummer	1
ISSN	2673-4591
DOI	https://doi.org/10.3390/engproc2024068016
Status	Udgivet - 2024
Udgivet eksternt	Ja

Adgang til dokumentet

10.3390/engproc2024068016

Biblioteksadgang?

Tjek tilgængelighed hos det Kgl. Bibliotek

Citationsformater

@article{2c69c328ae2340c89cc2b402c95d30ba,

title = "Modelling the Daily Concentration of Airborne Particles Using 1D Convolutional Neural Networks",

abstract = "This paper focuses on improving the prediction of the daily concentration of the pollutants, PM10 and nitrogen oxides (NO, NO2) in the air at urban monitoring sites using 1D convolutional neural networks (CNN). The results show that the 1D CNN model outperforms the other machine learning models (LSTM and Random Forest) in terms of the coefficients of determination and absolute errors.",

keywords = "1D CNN, air pollution, machine learning",

author = "Ivan Gudelj and Mario Lovri{\'c} and Nyarko, {Emmanuel Karlo}",

year = "2024",

doi = "10.3390/engproc2024068016",

language = "English",

volume = "68",

journal = "Engineering Proceedings",

issn = "2673-4591",

publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",

number = "1",

}

TY - JOUR

T1 - Modelling the Daily Concentration of Airborne Particles Using 1D Convolutional Neural Networks

AU - Gudelj, Ivan

AU - Lovrić, Mario

AU - Nyarko, Emmanuel Karlo

PY - 2024

Y1 - 2024

N2 - This paper focuses on improving the prediction of the daily concentration of the pollutants, PM10 and nitrogen oxides (NO, NO2) in the air at urban monitoring sites using 1D convolutional neural networks (CNN). The results show that the 1D CNN model outperforms the other machine learning models (LSTM and Random Forest) in terms of the coefficients of determination and absolute errors.

AB - This paper focuses on improving the prediction of the daily concentration of the pollutants, PM10 and nitrogen oxides (NO, NO2) in the air at urban monitoring sites using 1D convolutional neural networks (CNN). The results show that the 1D CNN model outperforms the other machine learning models (LSTM and Random Forest) in terms of the coefficients of determination and absolute errors.

KW - 1D CNN

KW - air pollution

KW - machine learning

U2 - 10.3390/engproc2024068016

DO - 10.3390/engproc2024068016

M3 - Journal article

SN - 2673-4591

VL - 68

JO - Engineering Proceedings

JF - Engineering Proceedings

IS - 1

ER -