Uncertainty-Aware Deep Learning Architectures for Highly Dynamic Air Quality Prediction

Forecasting air pollution is considered as an essential key for early warning and control management of air pollution, especially in emergency situations, where big amounts of pollutants are quickly released in the air, causing considerable damages. Predicting pollution in such situations is particu...

Full description

Bibliographic Details
Main Authors: Ichrak Mokhtari, Walid Bechkit, Herve Rivano, Mouloud Riadh Yaici
Format: Article
Language:English
Published: IEEE 2021-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/9328097/
id doaj-44910e6c7ffb44c2bdd18d79f08a919a
record_format Article
spelling doaj-44910e6c7ffb44c2bdd18d79f08a919a2021-03-30T15:14:51ZengIEEEIEEE Access2169-35362021-01-019147651477810.1109/ACCESS.2021.30524299328097Uncertainty-Aware Deep Learning Architectures for Highly Dynamic Air Quality PredictionIchrak Mokhtari0https://orcid.org/0000-0001-6993-9612Walid Bechkit1https://orcid.org/0000-0002-5438-4033Herve Rivano2https://orcid.org/0000-0001-6112-7468Mouloud Riadh Yaici3CITI Laboratory, INSA Lyon, Villeurbanne, FranceCITI Laboratory, INSA Lyon, Villeurbanne, FranceCITI Laboratory, INSA Lyon, Villeurbanne, FranceCITI Laboratory, INSA Lyon, Villeurbanne, FranceForecasting air pollution is considered as an essential key for early warning and control management of air pollution, especially in emergency situations, where big amounts of pollutants are quickly released in the air, causing considerable damages. Predicting pollution in such situations is particularly challenging due to the strong dynamic of the phenomenon and the various spatio-temporal factors affecting air pollution dispersion. In addition, providing uncertainty estimates of prediction makes the forecasting model more trustworthy, which helps decision-makers to take appropriate actions with more confidence regarding the pollution crisis. In this study, we propose a multi-point deep learning model based on convolutional long short term memory (ConvLSTM) for highly dynamic air quality forecasting. ConvLSTM architectures combines long short term memory (LSTM) and convolutional neural network (CNN), which allows to mine both temporal and spatial data features. In addition, uncertainty quantification methods were implemented on top of our model's architecture and their performances were further excavated. We conduct extensive experimental evaluations using a real and highly dynamic air pollution data set called Fusion Field Trial 2007 (FFT07). The results demonstrate the superiority of our proposed deep learning model in comparison to state-of-the-art methods including machine and deep learning techniques. Finally, we discuss the results of the uncertainty techniques and we derive insights.https://ieeexplore.ieee.org/document/9328097/Conv-LSTMspatio-temporel predictionhighly dynamic air qualityaccidental pollutant releaseuncertaintyFFT-07
collection DOAJ
language English
format Article
sources DOAJ
author Ichrak Mokhtari
Walid Bechkit
Herve Rivano
Mouloud Riadh Yaici
spellingShingle Ichrak Mokhtari
Walid Bechkit
Herve Rivano
Mouloud Riadh Yaici
Uncertainty-Aware Deep Learning Architectures for Highly Dynamic Air Quality Prediction
IEEE Access
Conv-LSTM
spatio-temporel prediction
highly dynamic air quality
accidental pollutant release
uncertainty
FFT-07
author_facet Ichrak Mokhtari
Walid Bechkit
Herve Rivano
Mouloud Riadh Yaici
author_sort Ichrak Mokhtari
title Uncertainty-Aware Deep Learning Architectures for Highly Dynamic Air Quality Prediction
title_short Uncertainty-Aware Deep Learning Architectures for Highly Dynamic Air Quality Prediction
title_full Uncertainty-Aware Deep Learning Architectures for Highly Dynamic Air Quality Prediction
title_fullStr Uncertainty-Aware Deep Learning Architectures for Highly Dynamic Air Quality Prediction
title_full_unstemmed Uncertainty-Aware Deep Learning Architectures for Highly Dynamic Air Quality Prediction
title_sort uncertainty-aware deep learning architectures for highly dynamic air quality prediction
publisher IEEE
series IEEE Access
issn 2169-3536
publishDate 2021-01-01
description Forecasting air pollution is considered as an essential key for early warning and control management of air pollution, especially in emergency situations, where big amounts of pollutants are quickly released in the air, causing considerable damages. Predicting pollution in such situations is particularly challenging due to the strong dynamic of the phenomenon and the various spatio-temporal factors affecting air pollution dispersion. In addition, providing uncertainty estimates of prediction makes the forecasting model more trustworthy, which helps decision-makers to take appropriate actions with more confidence regarding the pollution crisis. In this study, we propose a multi-point deep learning model based on convolutional long short term memory (ConvLSTM) for highly dynamic air quality forecasting. ConvLSTM architectures combines long short term memory (LSTM) and convolutional neural network (CNN), which allows to mine both temporal and spatial data features. In addition, uncertainty quantification methods were implemented on top of our model's architecture and their performances were further excavated. We conduct extensive experimental evaluations using a real and highly dynamic air pollution data set called Fusion Field Trial 2007 (FFT07). The results demonstrate the superiority of our proposed deep learning model in comparison to state-of-the-art methods including machine and deep learning techniques. Finally, we discuss the results of the uncertainty techniques and we derive insights.
topic Conv-LSTM
spatio-temporel prediction
highly dynamic air quality
accidental pollutant release
uncertainty
FFT-07
url https://ieeexplore.ieee.org/document/9328097/
work_keys_str_mv AT ichrakmokhtari uncertaintyawaredeeplearningarchitecturesforhighlydynamicairqualityprediction
AT walidbechkit uncertaintyawaredeeplearningarchitecturesforhighlydynamicairqualityprediction
AT herverivano uncertaintyawaredeeplearningarchitecturesforhighlydynamicairqualityprediction
AT mouloudriadhyaici uncertaintyawaredeeplearningarchitecturesforhighlydynamicairqualityprediction
_version_ 1724179737415254016