Comparing different methods for statistical modeling of particulate matter in Tehran, Iran

Mehdipour, Vahid; Stevenson, David S.; Memarianfard, M.; Sihag, Parveen

doi:10.1007/s11869-018-0615-z

Cited by 55 publications

(16 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Singh et al ( 2013 ) used an SVM model for predicting urban air quality in India where the RMSE value was 9.14 μgm −3 and 9.22 μgm −3 during testing and training period, respectively. However, in Tehran, Mehdipour et al ( 2018 ) experimented much lower RMSE value (0.0501 μgm −3 and 0.519 μgm −3 during testing and training respectively) using SVM models to predict PM 2.5 concentration.…”

Section: Resultsmentioning

confidence: 98%

“…Over-fitting was controlled in this study during the model execution. Generally, over-fitting occurs when the results of testing are greater than the validation (Mehdipour et al 2018 ). Singh et al ( 2013 ) used an SVM model for predicting urban air quality in India where the RMSE value was 9.14 μgm −3 and 9.22 μgm −3 during testing and training period, respectively.…”

Section: Resultsmentioning

confidence: 99%

“…The worst performance of GPR was found for PM 10 prediction in MS-2 (Table 2 ). A study in Tehran, i.e., Mehdipour et al ( 2018 ), used Bayesian network to predict PM 2.5 where the final RMSE value was 0.1077. Figure 1 represents the overall validation results of the models.…”

Section: Resultsmentioning

confidence: 99%

See 2 more Smart Citations

Applicability of machine learning in modeling of atmospheric particle pollution in Bangladesh

Shahriar

Kayes

Hasan

et al. 2020

Air Qual Atmos Health

View full text Add to dashboard Cite

Atmospheric particle pollution causes acute and chronic health effects. Predicting the concentrations of PM 2.5 and PM 10 , therefore, is a prerequisite to avoid the consequences and mitigate the complications. This research utilized the machine learning (ML) models such as linear-support vector machine (L-SVM), medium Gaussian-support vector machine (M-SVM), Gaussian process regression (GPR), artificial neural network (ANN), random forest regression (RFR), and a time series model namely PROPHET. Atmospheric NO X , SO 2 , CO, and O 3 , along with meteorological variables from Dhaka, Chattogram, Rajshahi, and Sylhet for the period of 2013 to 2019, were utilized as exploratory variables. Results showed that the overall performance of GPR performed better particularly for Dhaka in predicting the concentration of both PM 2.5 and PM 10 while ANN performed best in case of Chattogram and Sylhet for predicting PM 2.5. However, in terms of predicting PM 10 , M-SVM and RFR were selected respectively. Therefore, this study recommends utilizing "ensemble learning" models by combining several best models to advance application of ML in predicting pollutants' concentration in Bangladesh.

show abstract

Section: Resultsmentioning

confidence: 98%

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Applicability of machine learning in modeling of atmospheric particle pollution in Bangladesh

Shahriar

Kayes

Hasan

et al. 2020

Air Qual Atmos Health

View full text Add to dashboard Cite

show abstract

“…Some latest studies suggested successful application of soft computing techniques, viz. SVM, GPR, M5 tree and random forest regression to the field of groundwater hydrology (Singh et al 2017(Singh et al , 2019aAngelaki et al 2018;Sihag et al 2018a, b, c;Vand et al 2018;Sihag et al 2019b, c), water resources (Kumar et al 2018;Sepahvand et al 2019;Singh et al 2018a, b;Tiwari and Sihag 2018;Tiwari et al 2019) and engineering (Nain et al 2018(Nain et al , 2019Mehdipour et al 2018;Mohanty et al 2019). Keeping in view the importance of M5 tree and random forest regression techniques, the present research deals with the implementation of these techniques in an attempt to relate unsaturated hydraulic conductivity of the field data measured from 20 locations of Kurukshetra district, Haryana, with the soil physical properties.…”

Section: Introductionmentioning

confidence: 99%

Random forest, M5P and regression analysis to estimate the field unsaturated hydraulic conductivity

2019

Self Cite

View full text Add to dashboard Cite

Hydraulic conductivity of soil reveals its influencing role in the studies related to management of surface and subsurface flow, e.g. irrigation and drainage projects, and solute mass transport models. Direct measurements of hydraulic conductivity have many difficulties due to spatial variation of the property in the field. Pertaining to this problem, in this study, estimation models have been developed using machine learning methods (M5 tree model and random forest model) in an attempt to estimate the accurate values of unsaturated hydraulic conductivity related to basic soil properties (clay, silt and sand content, bulk density and moisture content). Data set was collected from the experimental measurements of cumulative infiltration using mini disc infiltrometer at the study area (Kurukshetra, India). A multivariate nonlinear regression (MNLR) relationship was derived, and the performance of this model was compared with the machine learning-based models. The evaluation of the results, based on statistical criteria (R 2 , RMSE, MAE), suggested that random forest regression model is superior in accurate estimations of the unsaturated hydraulic conductivity of field data relative to M5 model tree and MNLR.

show abstract

“…Air pollution models represent an important tool in environmental and epidemiological science (Liu et al 2019;Mehdipour et al 2018). Little attention is given to applying the modelling technique to investigate not only share of PM Urban population in percentage exposed to air pollutant concentrations above selected limit and target values according to a UE air-quality standards; 1 b WHO air quality guideline.…”

Section: Introductionmentioning

confidence: 99%

Air pollution in European countries and life expectancy—modelling with the use of neural network

Kolasa-Więcek

Suszanowicz

2019

Air Qual Atmos Health

View full text Add to dashboard Cite

The present paper discusses a novel methodology based on neural network to determine air pollutants' correlation with life expectancy in European countries. The models were developed using historical data from the period 1992-2016, for a set of 20 European countries. The subject of the analysis included the input variables of the following air pollutants: sulphur oxides, nitrogen oxides, carbon monoxide, particulate matters, polycyclic aromatic hydrocarbons and non-methane volatile organic compounds. Our main findings indicate that all the variables significantly affect life expectancy. Sensitivity of constructed neural networks to pollutants proved to be particularly important in the case of changes in the value of particulate matters, sulphur oxides and non-methane volatile organic compounds. The most frequent association was found for fine particle. Modelled courses of changes in the variable under study coincide with the actual data, which confirms that the proposed models generalize acquired knowledge well.

show abstract

Comparing different methods for statistical modeling of particulate matter in Tehran, Iran

Cited by 55 publications

References 57 publications

Applicability of machine learning in modeling of atmospheric particle pollution in Bangladesh

Applicability of machine learning in modeling of atmospheric particle pollution in Bangladesh

Random forest, M5P and regression analysis to estimate the field unsaturated hydraulic conductivity

Air pollution in European countries and life expectancy—modelling with the use of neural network

Contact Info

Product

Resources

About