The Explainable Potential of Coupling Metaheuristics-Optimized-XGBoost and SHAP in Revealing VOCs’ Environmental Fate
In this paper, we explore the computational capabilities of advanced modeling tools to reveal the factors that shape the observed benzene levels and behavior under different environmental conditions. The research was based on two-year hourly data concentrations of inorganic gaseous pollutants, parti...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-01-01
|
| Series: | Atmosphere |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2073-4433/14/1/109 |
| _version_ | 1797446300504948736 |
|---|---|
| author | Luka Jovanovic Gordana Jovanovic Mirjana Perisic Filip Alimpic Svetlana Stanisic Nebojsa Bacanin Miodrag Zivkovic Andreja Stojic |
| author_facet | Luka Jovanovic Gordana Jovanovic Mirjana Perisic Filip Alimpic Svetlana Stanisic Nebojsa Bacanin Miodrag Zivkovic Andreja Stojic |
| author_sort | Luka Jovanovic |
| collection | DOAJ |
| description | In this paper, we explore the computational capabilities of advanced modeling tools to reveal the factors that shape the observed benzene levels and behavior under different environmental conditions. The research was based on two-year hourly data concentrations of inorganic gaseous pollutants, particulate matter, benzene, toluene, m, p-xylenes, total nonmethane hydrocarbons, and meteorological parameters obtained from the Global Data Assimilation System. In order to determine the model that will be capable of achieving a superior level of performance, eight metaheuristics algorithms were tested for eXtreme Gradient Boosting optimization, while the relative SHapley Additive exPlanations values were used to estimate the relative importance of each pollutant level and meteorological parameter for the prediction of benzene concentrations. According to the results, benzene levels are mostly shaped by toluene and the finest aerosol fraction concentrations, in the environment governed by temperature, volumetric soil moisture content, and momentum flux direction, as well as by levels of total nonmethane hydrocarbons and total nitrogen oxide. The types of conditions which provided the environment for the impact of toluene, the finest aerosol, and temperature on benzene dynamics are distinguished and described. |
| first_indexed | 2024-03-09T13:38:20Z |
| format | Article |
| id | doaj.art-a1e20a96b93a4e86b489878c8752033a |
| institution | Directory Open Access Journal |
| issn | 2073-4433 |
| language | English |
| last_indexed | 2024-03-09T13:38:20Z |
| publishDate | 2023-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Atmosphere |
| spelling | doaj.art-a1e20a96b93a4e86b489878c8752033a2023-11-30T21:09:31ZengMDPI AGAtmosphere2073-44332023-01-0114110910.3390/atmos14010109The Explainable Potential of Coupling Metaheuristics-Optimized-XGBoost and SHAP in Revealing VOCs’ Environmental FateLuka Jovanovic0Gordana Jovanovic1Mirjana Perisic2Filip Alimpic3Svetlana Stanisic4Nebojsa Bacanin5Miodrag Zivkovic6Andreja Stojic7Faculty of Informatics and Computing, Singidunum University, Danijelova 32, 11010 Belgrade, SerbiaInstitute of Physics Belgrade, National Institute of the Republic of Serbia, University of Belgrade, Pregrevica 118, 11080 Belgrade, SerbiaInstitute of Physics Belgrade, National Institute of the Republic of Serbia, University of Belgrade, Pregrevica 118, 11080 Belgrade, SerbiaInstitute of Physics Belgrade, National Institute of the Republic of Serbia, University of Belgrade, Pregrevica 118, 11080 Belgrade, SerbiaEnvironment and Sustainable Development, Singidunum University, Danijelova 32, 11010 Belgrade, SerbiaFaculty of Informatics and Computing, Singidunum University, Danijelova 32, 11010 Belgrade, SerbiaFaculty of Informatics and Computing, Singidunum University, Danijelova 32, 11010 Belgrade, SerbiaInstitute of Physics Belgrade, National Institute of the Republic of Serbia, University of Belgrade, Pregrevica 118, 11080 Belgrade, SerbiaIn this paper, we explore the computational capabilities of advanced modeling tools to reveal the factors that shape the observed benzene levels and behavior under different environmental conditions. The research was based on two-year hourly data concentrations of inorganic gaseous pollutants, particulate matter, benzene, toluene, m, p-xylenes, total nonmethane hydrocarbons, and meteorological parameters obtained from the Global Data Assimilation System. In order to determine the model that will be capable of achieving a superior level of performance, eight metaheuristics algorithms were tested for eXtreme Gradient Boosting optimization, while the relative SHapley Additive exPlanations values were used to estimate the relative importance of each pollutant level and meteorological parameter for the prediction of benzene concentrations. According to the results, benzene levels are mostly shaped by toluene and the finest aerosol fraction concentrations, in the environment governed by temperature, volumetric soil moisture content, and momentum flux direction, as well as by levels of total nonmethane hydrocarbons and total nitrogen oxide. The types of conditions which provided the environment for the impact of toluene, the finest aerosol, and temperature on benzene dynamics are distinguished and described.https://www.mdpi.com/2073-4433/14/1/109machine learningextreme gradient boostingmetaheuristicsswarm intelligenceexplainable artificial intelligencevolatile organic compounds |
| spellingShingle | Luka Jovanovic Gordana Jovanovic Mirjana Perisic Filip Alimpic Svetlana Stanisic Nebojsa Bacanin Miodrag Zivkovic Andreja Stojic The Explainable Potential of Coupling Metaheuristics-Optimized-XGBoost and SHAP in Revealing VOCs’ Environmental Fate Atmosphere machine learning extreme gradient boosting metaheuristics swarm intelligence explainable artificial intelligence volatile organic compounds |
| title | The Explainable Potential of Coupling Metaheuristics-Optimized-XGBoost and SHAP in Revealing VOCs’ Environmental Fate |
| title_full | The Explainable Potential of Coupling Metaheuristics-Optimized-XGBoost and SHAP in Revealing VOCs’ Environmental Fate |
| title_fullStr | The Explainable Potential of Coupling Metaheuristics-Optimized-XGBoost and SHAP in Revealing VOCs’ Environmental Fate |
| title_full_unstemmed | The Explainable Potential of Coupling Metaheuristics-Optimized-XGBoost and SHAP in Revealing VOCs’ Environmental Fate |
| title_short | The Explainable Potential of Coupling Metaheuristics-Optimized-XGBoost and SHAP in Revealing VOCs’ Environmental Fate |
| title_sort | explainable potential of coupling metaheuristics optimized xgboost and shap in revealing vocs environmental fate |
| topic | machine learning extreme gradient boosting metaheuristics swarm intelligence explainable artificial intelligence volatile organic compounds |
| url | https://www.mdpi.com/2073-4433/14/1/109 |
| work_keys_str_mv | AT lukajovanovic theexplainablepotentialofcouplingmetaheuristicsoptimizedxgboostandshapinrevealingvocsenvironmentalfate AT gordanajovanovic theexplainablepotentialofcouplingmetaheuristicsoptimizedxgboostandshapinrevealingvocsenvironmentalfate AT mirjanaperisic theexplainablepotentialofcouplingmetaheuristicsoptimizedxgboostandshapinrevealingvocsenvironmentalfate AT filipalimpic theexplainablepotentialofcouplingmetaheuristicsoptimizedxgboostandshapinrevealingvocsenvironmentalfate AT svetlanastanisic theexplainablepotentialofcouplingmetaheuristicsoptimizedxgboostandshapinrevealingvocsenvironmentalfate AT nebojsabacanin theexplainablepotentialofcouplingmetaheuristicsoptimizedxgboostandshapinrevealingvocsenvironmentalfate AT miodragzivkovic theexplainablepotentialofcouplingmetaheuristicsoptimizedxgboostandshapinrevealingvocsenvironmentalfate AT andrejastojic theexplainablepotentialofcouplingmetaheuristicsoptimizedxgboostandshapinrevealingvocsenvironmentalfate AT lukajovanovic explainablepotentialofcouplingmetaheuristicsoptimizedxgboostandshapinrevealingvocsenvironmentalfate AT gordanajovanovic explainablepotentialofcouplingmetaheuristicsoptimizedxgboostandshapinrevealingvocsenvironmentalfate AT mirjanaperisic explainablepotentialofcouplingmetaheuristicsoptimizedxgboostandshapinrevealingvocsenvironmentalfate AT filipalimpic explainablepotentialofcouplingmetaheuristicsoptimizedxgboostandshapinrevealingvocsenvironmentalfate AT svetlanastanisic explainablepotentialofcouplingmetaheuristicsoptimizedxgboostandshapinrevealingvocsenvironmentalfate AT nebojsabacanin explainablepotentialofcouplingmetaheuristicsoptimizedxgboostandshapinrevealingvocsenvironmentalfate AT miodragzivkovic explainablepotentialofcouplingmetaheuristicsoptimizedxgboostandshapinrevealingvocsenvironmentalfate AT andrejastojic explainablepotentialofcouplingmetaheuristicsoptimizedxgboostandshapinrevealingvocsenvironmentalfate |