Predicting next hour fine particulate matter (PM2.5) in the Istanbul Metropolitan City using deep learning algorithms with time windowing strategy. Eren, B., Aksangür, İ., & Erden, C. Urban Climate, 48:101418, March, 2023. ![Predicting next hour fine particulate matter (PM2.5) in the Istanbul Metropolitan City using deep learning algorithms with time windowing strategy [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex 1 download Poor air quality has various detrimental physical and mental effects on human health and quality of life. In particular, PM2.5 air pollution has been associated with cardiovascular and respiratory problems. Therefore, air quality management is an essential issue for densely populated cities to reduce or prevent the adverse effects of air pollution. Considering this, reliable models for predicting pollution levels for pollutants like PM2.5 are critical tools for decision-making. For this purpose, this study presents three kinds of deep learning (DL) algorithms (LSTM, RNN, and GRU) that utilize a time-windowing strategy to predict the hourly concentration of PM2.5 in the Istanbul metropolitan. The models were trained and tested using large data sets that envelope air quality parameters (PM2.5, SO2, NO, NO2, NOX, and O3) and meteorological factors (temperature, wind speed, relative humidity, and air pressure) for about five years. The experimental results demonstrate that the LSTM+LSTM model performs significantly better with an R2 of 0.98 and 0.97 at the significance level (p \textless 0.05) for training and test sets compared to other deep learning algorithms. In addition, data for one year from several stations located in nine different districts of Istanbul were used to evaluate the proposed model's generalization ability. As a result, the proposed LSTM+LSTM model has a good generalization ability with an R2 accuracy rate of 0.90 (p \textless 0.05) and above for all stations and can be used for non-linear, non-stationary multidimensional time series data. Furthermore, the results were compared to other studies in the literature; it was found that the proposed LSTM+LSTM model performed better in predicting PM2.5 concentrations.
@article{eren_predicting_2023,
title = {Predicting next hour fine particulate matter ({PM2}.5) in the {Istanbul} {Metropolitan} {City} using deep learning algorithms with time windowing strategy},
volume = {48},
copyright = {All rights reserved},
issn = {2212-0955},
url = {https://doi.org/10.1016/j.uclim.2023.101418},
doi = {10.1016/j.uclim.2023.101418},
abstract = {Poor air quality has various detrimental physical and mental effects on human health and quality of life. In particular, PM2.5 air pollution has been associated with cardiovascular and respiratory problems. Therefore, air quality management is an essential issue for densely populated cities to reduce or prevent the adverse effects of air pollution. Considering this, reliable models for predicting pollution levels for pollutants like PM2.5 are critical tools for decision-making. For this purpose, this study presents three kinds of deep learning (DL) algorithms (LSTM, RNN, and GRU) that utilize a time-windowing strategy to predict the hourly concentration of PM2.5 in the Istanbul metropolitan. The models were trained and tested using large data sets that envelope air quality parameters (PM2.5, SO2, NO, NO2, NOX, and O3) and meteorological factors (temperature, wind speed, relative humidity, and air pressure) for about five years. The experimental results demonstrate that the LSTM+LSTM model performs significantly better with an R2 of 0.98 and 0.97 at the significance level (p {\textless} 0.05) for training and test sets compared to other deep learning algorithms. In addition, data for one year from several stations located in nine different districts of Istanbul were used to evaluate the proposed model's generalization ability. As a result, the proposed LSTM+LSTM model has a good generalization ability with an R2 accuracy rate of 0.90 (p {\textless} 0.05) and above for all stations and can be used for non-linear, non-stationary multidimensional time series data. Furthermore, the results were compared to other studies in the literature; it was found that the proposed LSTM+LSTM model performed better in predicting PM2.5 concentrations.},
journal = {Urban Climate},
author = {Eren, Beytullah and Aksangür, İpek and Erden, Caner},
month = mar,
year = {2023},
keywords = {Deep learning, Fine particulate matter (PM), Gated recurrent unit (GRU), Long-short term memory (LSTM), Recurrent neural network (RNN), Time windowing},
pages = {101418},
}
Downloads: 1
{"_id":"unGXMfhvJJAJHbxMy","bibbaseid":"eren-aksangr-erden-predictingnexthourfineparticulatematterpm25intheistanbulmetropolitancityusingdeeplearningalgorithmswithtimewindowingstrategy-2023","author_short":["Eren, B.","Aksangür, İ.","Erden, C."],"bibdata":{"bibtype":"article","type":"article","title":"Predicting next hour fine particulate matter (PM2.5) in the Istanbul Metropolitan City using deep learning algorithms with time windowing strategy","volume":"48","copyright":"All rights reserved","issn":"2212-0955","url":"https://doi.org/10.1016/j.uclim.2023.101418","doi":"10.1016/j.uclim.2023.101418","abstract":"Poor air quality has various detrimental physical and mental effects on human health and quality of life. In particular, PM2.5 air pollution has been associated with cardiovascular and respiratory problems. Therefore, air quality management is an essential issue for densely populated cities to reduce or prevent the adverse effects of air pollution. Considering this, reliable models for predicting pollution levels for pollutants like PM2.5 are critical tools for decision-making. For this purpose, this study presents three kinds of deep learning (DL) algorithms (LSTM, RNN, and GRU) that utilize a time-windowing strategy to predict the hourly concentration of PM2.5 in the Istanbul metropolitan. The models were trained and tested using large data sets that envelope air quality parameters (PM2.5, SO2, NO, NO2, NOX, and O3) and meteorological factors (temperature, wind speed, relative humidity, and air pressure) for about five years. The experimental results demonstrate that the LSTM+LSTM model performs significantly better with an R2 of 0.98 and 0.97 at the significance level (p \\textless 0.05) for training and test sets compared to other deep learning algorithms. In addition, data for one year from several stations located in nine different districts of Istanbul were used to evaluate the proposed model's generalization ability. As a result, the proposed LSTM+LSTM model has a good generalization ability with an R2 accuracy rate of 0.90 (p \\textless 0.05) and above for all stations and can be used for non-linear, non-stationary multidimensional time series data. Furthermore, the results were compared to other studies in the literature; it was found that the proposed LSTM+LSTM model performed better in predicting PM2.5 concentrations.","journal":"Urban Climate","author":[{"propositions":[],"lastnames":["Eren"],"firstnames":["Beytullah"],"suffixes":[]},{"propositions":[],"lastnames":["Aksangür"],"firstnames":["İpek"],"suffixes":[]},{"propositions":[],"lastnames":["Erden"],"firstnames":["Caner"],"suffixes":[]}],"month":"March","year":"2023","keywords":"Deep learning, Fine particulate matter (PM), Gated recurrent unit (GRU), Long-short term memory (LSTM), Recurrent neural network (RNN), Time windowing","pages":"101418","bibtex":"@article{eren_predicting_2023,\n\ttitle = {Predicting next hour fine particulate matter ({PM2}.5) in the {Istanbul} {Metropolitan} {City} using deep learning algorithms with time windowing strategy},\n\tvolume = {48},\n\tcopyright = {All rights reserved},\n\tissn = {2212-0955},\n\turl = {https://doi.org/10.1016/j.uclim.2023.101418},\n\tdoi = {10.1016/j.uclim.2023.101418},\n\tabstract = {Poor air quality has various detrimental physical and mental effects on human health and quality of life. In particular, PM2.5 air pollution has been associated with cardiovascular and respiratory problems. Therefore, air quality management is an essential issue for densely populated cities to reduce or prevent the adverse effects of air pollution. Considering this, reliable models for predicting pollution levels for pollutants like PM2.5 are critical tools for decision-making. For this purpose, this study presents three kinds of deep learning (DL) algorithms (LSTM, RNN, and GRU) that utilize a time-windowing strategy to predict the hourly concentration of PM2.5 in the Istanbul metropolitan. The models were trained and tested using large data sets that envelope air quality parameters (PM2.5, SO2, NO, NO2, NOX, and O3) and meteorological factors (temperature, wind speed, relative humidity, and air pressure) for about five years. The experimental results demonstrate that the LSTM+LSTM model performs significantly better with an R2 of 0.98 and 0.97 at the significance level (p {\\textless} 0.05) for training and test sets compared to other deep learning algorithms. In addition, data for one year from several stations located in nine different districts of Istanbul were used to evaluate the proposed model's generalization ability. As a result, the proposed LSTM+LSTM model has a good generalization ability with an R2 accuracy rate of 0.90 (p {\\textless} 0.05) and above for all stations and can be used for non-linear, non-stationary multidimensional time series data. Furthermore, the results were compared to other studies in the literature; it was found that the proposed LSTM+LSTM model performed better in predicting PM2.5 concentrations.},\n\tjournal = {Urban Climate},\n\tauthor = {Eren, Beytullah and Aksangür, İpek and Erden, Caner},\n\tmonth = mar,\n\tyear = {2023},\n\tkeywords = {Deep learning, Fine particulate matter (PM), Gated recurrent unit (GRU), Long-short term memory (LSTM), Recurrent neural network (RNN), Time windowing},\n\tpages = {101418},\n}\n\n","author_short":["Eren, B.","Aksangür, İ.","Erden, C."],"key":"eren_predicting_2023","id":"eren_predicting_2023","bibbaseid":"eren-aksangr-erden-predictingnexthourfineparticulatematterpm25intheistanbulmetropolitancityusingdeeplearningalgorithmswithtimewindowingstrategy-2023","role":"author","urls":{"Paper":"https://doi.org/10.1016/j.uclim.2023.101418"},"keyword":["Deep learning","Fine particulate matter (PM)","Gated recurrent unit (GRU)","Long-short term memory (LSTM)","Recurrent neural network (RNN)","Time windowing"],"metadata":{"authorlinks":{}},"downloads":1},"bibtype":"article","biburl":"https://api.zotero.org/users/2824292/collections/QT62GLCD/items?key=ayPetSSPAL2MSiGGBpMUXrTE&format=bibtex&limit=100","dataSources":["M2aEWDBqKoSjSrgh9","M9EojWM4w57Ky8uX4","mue4g8pWxLCtg5mmp","jkcBw6uJf2fF8Ac5u"],"keywords":["deep learning","fine particulate matter (pm)","gated recurrent unit (gru)","long-short term memory (lstm)","recurrent neural network (rnn)","time windowing"],"search_terms":["predicting","next","hour","fine","particulate","matter","pm2","istanbul","metropolitan","city","using","deep","learning","algorithms","time","windowing","strategy","eren","aksangür","erden"],"title":"Predicting next hour fine particulate matter (PM2.5) in the Istanbul Metropolitan City using deep learning algorithms with time windowing strategy","year":2023,"downloads":1}