Evaluation of Artificial Neural Network Model for Predicting Nitrogen Oxides (NOₓ) Concentration

Nitrogen Oxides (NOₓ) are air pollutants that require serious attention due to their potential negative impacts on human health, the environment, and the economy. This research is crucial to provide accurate predictive models of NOₓ concentration, which can serve as a foundation for decision-making...

Full description

Saved in:
Bibliographic Details
Main Authors: Muhammad Farrih Mahabbataka Arsyada, Raras Tyasnurita
Format: Article
Language:Indonesian
Published: Islamic University of Indragiri 2025-05-01
Series:Sistemasi: Jurnal Sistem Informasi
Subjects:
Online Access:https://sistemasi.ftik.unisi.ac.id/index.php/stmsi/article/view/4371
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849222093768491008
author Muhammad Farrih Mahabbataka Arsyada
Raras Tyasnurita
author_facet Muhammad Farrih Mahabbataka Arsyada
Raras Tyasnurita
author_sort Muhammad Farrih Mahabbataka Arsyada
collection DOAJ
description Nitrogen Oxides (NOₓ) are air pollutants that require serious attention due to their potential negative impacts on human health, the environment, and the economy. This research is crucial to provide accurate predictive models of NOₓ concentration, which can serve as a foundation for decision-making and effective air pollution mitigation measures. The objective of this study is to evaluate several artificial neural network (ANN) models to determine the most effective model for accurately predicting NOₓ concentrations. One of the methods used for predicting air pollution data, such as NOₓ, is artificial neural networks (ANN). In this study, four ANN models were constructed and evaluated: Feed Forward Neural Network (FNN), Time Lagged Neural Network (TLNN), Seasonal Artificial Neural Network (SANN), and Long Short-Term Memory (LSTM). The models predict NOₓ concentration using data from the air quality dataset provided by the UCI Machine Learning Repository. Testing results indicate that the LSTM model performs best, achieving the lowest error value, characterized by 24 input nodes, three hidden nodes, one output node, and 300 training epochs. The RMSE values for LSTM, FNN, TLNN, and SANN are 57.3, 62.8, 64, and 89, respectively.
format Article
id doaj-art-90f64b6485d74ae084867a0ec6aebde6
institution Kabale University
issn 2302-8149
2540-9719
language Indonesian
publishDate 2025-05-01
publisher Islamic University of Indragiri
record_format Article
series Sistemasi: Jurnal Sistem Informasi
spelling doaj-art-90f64b6485d74ae084867a0ec6aebde62025-08-26T08:05:46ZindIslamic University of IndragiriSistemasi: Jurnal Sistem Informasi2302-81492540-97192025-05-011431001101310.32520/stmsi.v14i3.43711066Evaluation of Artificial Neural Network Model for Predicting Nitrogen Oxides (NOₓ) ConcentrationMuhammad Farrih Mahabbataka Arsyada0Raras Tyasnurita1Sepuluh Nopember Institute of TechnologySepuluh Nopember Institute of TechnologyNitrogen Oxides (NOₓ) are air pollutants that require serious attention due to their potential negative impacts on human health, the environment, and the economy. This research is crucial to provide accurate predictive models of NOₓ concentration, which can serve as a foundation for decision-making and effective air pollution mitigation measures. The objective of this study is to evaluate several artificial neural network (ANN) models to determine the most effective model for accurately predicting NOₓ concentrations. One of the methods used for predicting air pollution data, such as NOₓ, is artificial neural networks (ANN). In this study, four ANN models were constructed and evaluated: Feed Forward Neural Network (FNN), Time Lagged Neural Network (TLNN), Seasonal Artificial Neural Network (SANN), and Long Short-Term Memory (LSTM). The models predict NOₓ concentration using data from the air quality dataset provided by the UCI Machine Learning Repository. Testing results indicate that the LSTM model performs best, achieving the lowest error value, characterized by 24 input nodes, three hidden nodes, one output node, and 300 training epochs. The RMSE values for LSTM, FNN, TLNN, and SANN are 57.3, 62.8, 64, and 89, respectively.https://sistemasi.ftik.unisi.ac.id/index.php/stmsi/article/view/4371lstmartificial neural networksair pollutionnoₓ concentrationpredictive model
spellingShingle Muhammad Farrih Mahabbataka Arsyada
Raras Tyasnurita
Evaluation of Artificial Neural Network Model for Predicting Nitrogen Oxides (NOₓ) Concentration
Sistemasi: Jurnal Sistem Informasi
lstm
artificial neural networks
air pollution
noₓ concentration
predictive model
title Evaluation of Artificial Neural Network Model for Predicting Nitrogen Oxides (NOₓ) Concentration
title_full Evaluation of Artificial Neural Network Model for Predicting Nitrogen Oxides (NOₓ) Concentration
title_fullStr Evaluation of Artificial Neural Network Model for Predicting Nitrogen Oxides (NOₓ) Concentration
title_full_unstemmed Evaluation of Artificial Neural Network Model for Predicting Nitrogen Oxides (NOₓ) Concentration
title_short Evaluation of Artificial Neural Network Model for Predicting Nitrogen Oxides (NOₓ) Concentration
title_sort evaluation of artificial neural network model for predicting nitrogen oxides noₓ concentration
topic lstm
artificial neural networks
air pollution
noₓ concentration
predictive model
url https://sistemasi.ftik.unisi.ac.id/index.php/stmsi/article/view/4371
work_keys_str_mv AT muhammadfarrihmahabbatakaarsyada evaluationofartificialneuralnetworkmodelforpredictingnitrogenoxidesnoxconcentration
AT rarastyasnurita evaluationofartificialneuralnetworkmodelforpredictingnitrogenoxidesnoxconcentration