Air Pollution Modelling by Machine Learning Methods
Precise environmental modelling of pollutants distributions represents a key factor for addresing the issue of urban air pollution. Nowadays, urban air pollution monitoring is primarily carried out by employing sparse networks of spatially distributed fixed stations. The work in this paper aims at i...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-11-01
|
| Series: | Modelling |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2673-3951/2/4/35 |
| _version_ | 1797502232852168704 |
|---|---|
| author | Petra Vidnerová Roman Neruda |
| author_facet | Petra Vidnerová Roman Neruda |
| author_sort | Petra Vidnerová |
| collection | DOAJ |
| description | Precise environmental modelling of pollutants distributions represents a key factor for addresing the issue of urban air pollution. Nowadays, urban air pollution monitoring is primarily carried out by employing sparse networks of spatially distributed fixed stations. The work in this paper aims at improving the situation by utilizing machine learning models to process the outputs of multi-sensor devices that are small, cheap, albeit less reliable, thus a massive urban deployment of those devices is possible. The main contribution of the paper is the design of a mathematical model providing sensor fusion to extract the information and transform it into the desired pollutant concentrations. Multi-sensor outputs are used as input information for a particular machine learning model trained to produce the CO, NO2, and NOx concentration estimates. Several state-of-the-art machine learning methods, including original algorithms proposed by the authors, are utilized in this study: kernel methods, regularization networks, regularization networks with composite kernels, and deep neural networks. All methods are augmented with a proper hyper-parameter search to achieve the optimal performance for each model. All the methods considered achieved vital results, deep neural networks exhibited the best generalization ability, and regularization networks with product kernels achieved the best fitting of the training set. |
| first_indexed | 2024-03-10T03:30:01Z |
| format | Article |
| id | doaj.art-27ea2e7c6e53439fbd297de64982504a |
| institution | Directory Open Access Journal |
| issn | 2673-3951 |
| language | English |
| last_indexed | 2024-03-10T03:30:01Z |
| publishDate | 2021-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Modelling |
| spelling | doaj.art-27ea2e7c6e53439fbd297de64982504a2023-11-23T09:43:10ZengMDPI AGModelling2673-39512021-11-012465967410.3390/modelling2040035Air Pollution Modelling by Machine Learning MethodsPetra Vidnerová0Roman Neruda1The Czech Academy of Sciences, Institute of Computer Science, 182 07 Prague, Czech RepublicThe Czech Academy of Sciences, Institute of Computer Science, 182 07 Prague, Czech RepublicPrecise environmental modelling of pollutants distributions represents a key factor for addresing the issue of urban air pollution. Nowadays, urban air pollution monitoring is primarily carried out by employing sparse networks of spatially distributed fixed stations. The work in this paper aims at improving the situation by utilizing machine learning models to process the outputs of multi-sensor devices that are small, cheap, albeit less reliable, thus a massive urban deployment of those devices is possible. The main contribution of the paper is the design of a mathematical model providing sensor fusion to extract the information and transform it into the desired pollutant concentrations. Multi-sensor outputs are used as input information for a particular machine learning model trained to produce the CO, NO2, and NOx concentration estimates. Several state-of-the-art machine learning methods, including original algorithms proposed by the authors, are utilized in this study: kernel methods, regularization networks, regularization networks with composite kernels, and deep neural networks. All methods are augmented with a proper hyper-parameter search to achieve the optimal performance for each model. All the methods considered achieved vital results, deep neural networks exhibited the best generalization ability, and regularization networks with product kernels achieved the best fitting of the training set.https://www.mdpi.com/2673-3951/2/4/35machine learningair pollutionsensorsdeep neural networksregularization networks |
| spellingShingle | Petra Vidnerová Roman Neruda Air Pollution Modelling by Machine Learning Methods Modelling machine learning air pollution sensors deep neural networks regularization networks |
| title | Air Pollution Modelling by Machine Learning Methods |
| title_full | Air Pollution Modelling by Machine Learning Methods |
| title_fullStr | Air Pollution Modelling by Machine Learning Methods |
| title_full_unstemmed | Air Pollution Modelling by Machine Learning Methods |
| title_short | Air Pollution Modelling by Machine Learning Methods |
| title_sort | air pollution modelling by machine learning methods |
| topic | machine learning air pollution sensors deep neural networks regularization networks |
| url | https://www.mdpi.com/2673-3951/2/4/35 |
| work_keys_str_mv | AT petravidnerova airpollutionmodellingbymachinelearningmethods AT romanneruda airpollutionmodellingbymachinelearningmethods |