Powered by OpenAIRE graph
Found an issue? Give us feedback
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/ IEEE Accessarrow_drop_down
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
IEEE Access
Article . 2023 . Peer-reviewed
License: CC BY
Data sources: Crossref
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
IEEE Access
Article . 2023
Data sources: DOAJ
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
ZENODO
Other literature type . 2023
License: CC BY
Data sources: ZENODO
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
NILU Brage
Article . 2023
Data sources: NILU Brage
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
https://dx.doi.org/10.60692/sv...
Other literature type . 2023
Data sources: Datacite
https://dx.doi.org/10.60692/50...
Other literature type . 2023
Data sources: Datacite
DBLP
Article . 2023
Data sources: DBLP
versions View all 10 versions
addClaim

Comparative Analysis of Deep Learning and Statistical Models for Air Pollutants Prediction in Urban Areas

التحليل المقارن للتعلم العميق والنماذج الإحصائية للتنبؤ بملوثات الهواء في المناطق الحضرية
Authors: Fareena Naz; Conor McCann; Muhammad Fahim; Tuan-Vu Cao; Ruth F. Hunter; Nguyen Trung Viet; Long D. Nguyen; +1 Authors

Comparative Analysis of Deep Learning and Statistical Models for Air Pollutants Prediction in Urban Areas

Abstract

La croissance rapide de l'urbanisation et de l'industrialisation entraîne une augmentation de la pollution de l'air et une mauvaise qualité de l'air. En raison de ses effets néfastes sur l'environnement naturel et la santé humaine, elle a été déclarée « urgence silencieuse de santé publique ». Pour faire face à ce défi mondial, il est important que les parties prenantes prennent les mesures nécessaires pour prévoir avec précision la pollution de l'air. Au cours des dernières années, les modèles de prévision basés sur l'apprentissage en profondeur sont prometteurs pour une prévision de la qualité de l'air plus efficace et efficiente que d'autres approches. Dans cette étude, nous avons effectué une analyse comparative de divers modèles de prévision en une seule étape basés sur l'apprentissage en profondeur, tels que mémoire à court terme (LSTM), unité récurrente fermée (GRU) et un modèle statistique pour prédire cinq polluants atmosphériques, à savoir le dioxyde d'azote (NO 2 ), l'ozone (O 3), le dioxyde de soufre (SO 2) et les particules (PM2,5 et PM10). Pour l'évaluation empirique, nous avons utilisé un ensemble de données accessibles au public recueillies en Irlande du Nord, à l'aide d'une station de surveillance de la qualité de l'air située dans le centre-ville de Belfast. Elle mesure la concentration de polluants atmosphériques. La performance des modèles de prévision est évaluée en fonction de trois paramètres de performance : (a) racine erreur quadratique moyenne (RMSE), (b) erreur absolue moyenne (MAE) et (c) R-carré (R 2 ). Le résultat montre que les modèles d'apprentissage profond ont systématiquement obtenu le moins de RMSE par rapport aux modèles statistiques avec une valeur de 0,59. En outre, le modèle d'apprentissage profond s'avère également avoir le score R 2 le plus élevé de 0,856.

El rápido crecimiento de la urbanización y la industrialización conduce a un aumento de la contaminación del aire y a una mala calidad del aire. Debido a sus efectos adversos sobre el medio ambiente natural y la salud humana, se ha declarado una "emergencia silenciosa de salud pública". Para hacer frente a este desafío global, es importante que las partes interesadas tomen las medidas necesarias para predecir con precisión la contaminación del aire. En los últimos años, los modelos de pronóstico basados en el aprendizaje profundo son prometedores para pronosticar la calidad del aire de manera más efectiva y eficiente que otros enfoques. En este estudio, hicimos un análisis comparativo de varios modelos de pronóstico de un solo paso basados en el aprendizaje profundo, como los modelos memoria a corto plazo (LSTM), unidad recurrente cerrada (Gru) y un modelo estadístico para predecir cinco contaminantes atmosféricos, a saber, dióxido de nitrógeno (NO 2 ), ozono (O 3 ), dióxido de azufre (SO 2) y materia particulada (PM2.5 y PM10). Para la evaluación empírica, utilizamos un conjunto de datos disponibles públicamente recopilados en Irlanda del Norte, utilizando una estación de monitoreo de la calidad del aire situada en el centro de la ciudad de Belfast. Mide la concentración de contaminantes atmosféricos. El rendimiento de los modelos de pronóstico se evalúa en función de tres métricas de rendimiento: (a) raíz error cuadrático medio (RMSE), (b) error absoluto medio (MAE) y (c) R-cuadrado (R 2 ). El resultado muestra que los modelos de aprendizaje profundo lograron consistentemente el menor RMSE en comparación con los modelos estadísticos con un valor de 0.59. Además, el modelo de aprendizaje profundo también tiene la puntuación R 2 más alta de 0.856.

Rapid growth in urbanization and industrialization leads to an increase in air pollution and poor air quality.Because of its adverse effects on the natural environment and human health, it's been declared a "silent public health emergency".To deal with this global challenge, accurate prediction of air pollution is important for stakeholders to take required actions.In recent years, deep learning-based forecasting models show promise for more effective and efficient forecasting of air quality than other approaches.In this study, we made a comparative analysis of various deep learning-based single-step forecasting models such as long short term memory (LSTM), gated recurrent unit (GRU), and a statistical model to predict five air pollutants namely Nitrogen Dioxide (NO 2 ), Ozone (O 3 ), Sulphur Dioxide (SO 2 ), and Particulate Matter (PM2.5, and PM10).For empirical evaluation, we used a publicly available dataset collected in Northern Ireland, using an air quality monitoring station situated in Belfast city centre.It measures the concentration of air pollutants.The performance of forecasting models is evaluated based on three performance metrics: (a) root mean square error (RMSE), (b) mean absolute error (MAE) and (c) R-squared (R 2 ).The result shows that deep learning models consistently achieved the least RMSE compared to the statistical models with a value of 0.59.In addition, the deep learning model is also found to have the highest R 2 score of 0.856.

النمو السريع في التحضر والتصنيع يؤدي إلى زيادة في تلوث الهواء وسوء جودة الهواء. بسبب آثاره الضارة على البيئة الطبيعية وصحة الإنسان، تم إعلانه "حالة طوارئ صحية عامة صامتة". للتعامل مع هذا التحدي العالمي، يعد التنبؤ الدقيق بتلوث الهواء أمرًا مهمًا لأصحاب المصلحة لاتخاذ الإجراءات المطلوبة. في السنوات الأخيرة، تظهر نماذج التنبؤ القائمة على التعلم العميق توقعات أكثر فعالية وكفاءة لجودة الهواء من النهج الأخرى. في هذه الدراسة، أجرينا تحليلًا مقارنًا لمختلف نماذج التنبؤ ذات الخطوة الواحدة القائمة على التعلم العميق مثل الذاكرة قصيرة المدى (LSTM)، والوحدة المتكررة المسورة (GRU)، ونموذج إحصائي للتنبؤ بخمسة ملوثات هواء وهي ثاني أكسيد النيتروجين (NO 2 )، والأوزون (O 3 )، وثاني أكسيد الكبريت (SO 2 )، والمادة الجسيمية (PM2.5، و PM10). للتقييم التجريبي، استخدمنا مجموعة بيانات متاحة للجمهور تم جمعها في أيرلندا الشمالية، باستخدام محطة مراقبة جودة الهواء الموجودة في وسط مدينة بلفاست. يقيس تركيز ملوثات الهواء. يتم تقييم أداء نماذج التنبؤ بناءً على ثلاثة مقاييس للأداء: (أ) الجذر متوسط الخطأ المربع (RMSE)، (ب) متوسط الخطأ المطلق (MAE) و (ج) R - squared (R 2 ). تظهر النتيجة أن نماذج التعلم العميق حققت باستمرار أقل RMSE مقارنة بالنماذج الإحصائية بقيمة 0.59. بالإضافة إلى ذلك، وجد أيضًا أن نموذج التعلم العميق لديه أعلى درجة R 2 تبلغ 0.856.

Countries
United Kingdom, Norway
Keywords

Artificial intelligence, Environmental Engineering, 330, 550, Health, Toxicology and Mutagenesis, Air pollution, FOS: Mechanical engineering, Organic chemistry, Estimating Vehicle Fuel Consumption and Emissions, Health Effects of Air Pollution, Pollutant, Environmental science, Low-Cost Air Quality Monitoring Systems, /dk/atira/pure/sustainabledevelopmentgoals/sustainable_cities_and_communities, Engineering, Meteorology, Machine learning, FOS: Mathematics, /dk/atira/pure/sustainabledevelopmentgoals/good_health_and_well_being; name=SDG 3 - Good Health and Well-being, Geography, Statistics, FOS: Environmental engineering, deep learning, Air Quality Monitoring, Deep learning, predictive models, Air quality index, Computer science, name=SDG 3 - Good Health and Well-being, name=SDG 11 - Sustainable Cities and Communities, TK1-9971, Particulates, Chemistry, machine learning, Air quality, Environmental Science, Physical Sciences, Automotive Engineering, statistical methods, /dk/atira/pure/sustainabledevelopmentgoals/good_health_and_well_being, Mean squared error, Electrical engineering. Electronics. Nuclear engineering, /dk/atira/pure/sustainabledevelopmentgoals/sustainable_cities_and_communities; name=SDG 11 - Sustainable Cities and Communities, Mathematics, Forecasting, Air pollutants

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    10
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Top 10%
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Top 10%
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Top 10%
Powered by OpenAIRE graph
Found an issue? Give us feedback
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
10
Top 10%
Top 10%
Top 10%
Green
gold