Review on air pollution of Delhi zone using machine learning algorithm

Sinha, Anurag; Singh, Shubham

doi:10.18502/japh.v5i4.6446

Cited by 2 publications

(4 citation statements)

References 28 publications

(35 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…XGBoost and Random Forest scored the highest for winter (R 2 = 0.75), while BD-LSTM was best in the summer and post-monsoon seasons (R 2 = 0.77 and 0.82). In previous research, the highest R 2 values for ozone have similarly been achieved with non-linear machine learning methods, including R 2 = 0.49 with Neural Power Networking [40]; R 2 = 0.72 with Random Forest [41]; R 2 = 0.84 with Boosted Decision Tree Regression [14]; and R 2 = 0.66 with a Multi-Layer Perceptron network [31]. Of course, these studies are not directly comparable because they were carried out at different locations with different data sets.…”

Section: Discussionmentioning

confidence: 77%

“…SVM is a commonly used machine learning method for studies based in Delhi [37,38], perhaps due to its robustness to outliers and relatively flexible implementation. Studies by Sinha et al [39,40] compared several machine learning algorithms for the daily prediction of several pollutants in Delhi. Shukla et al [41] tested linear regression and Random Forest regression for the prediction of the pollutants NO, NO 2 , and O 3 , in which site-specific predictions using Random Forest had the best results.…”

Section: Introductionmentioning

confidence: 99%

“…Eight machine learning methods were analyzed and tuned for optimal performance in this study, as compared to 2-3 methods in most other studies, offering more robust analyses for comparing models for real-world implementation [12,26,31,41]. For ozone prediction in Delhi in particular, this work provides a useful comparison with previous work [26,40,41]. XGBoost, a relatively new algorithm that has not been as widely used in pollutant forecasting, is one of the methods tested here.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Comparison of Machine Learning Methods to Forecast Tropospheric Ozone Levels in Delhi

Juarez¹,

Petersen

2021

Atmosphere

View full text Add to dashboard Cite

Ground-level ozone is a pollutant that is harmful to urban populations, particularly in developing countries where it is present in significant quantities. It greatly increases the risk of heart and lung diseases and harms agricultural crops. This study hypothesized that, as a secondary pollutant, ground-level ozone is amenable to 24 h forecasting based on measurements of weather conditions and primary pollutants such as nitrogen oxides and volatile organic compounds. We developed software to analyze hourly records of 12 air pollutants and 5 weather variables over the course of one year in Delhi, India. To determine the best predictive model, eight machine learning algorithms were tuned, trained, tested, and compared using cross-validation with hourly data for a full year. The algorithms, ranked by R2 values, were XGBoost (0.61), Random Forest (0.61), K-Nearest Neighbor Regression (0.55), Support Vector Regression (0.48), Decision Trees (0.43), AdaBoost (0.39), and linear regression (0.39). When trained by separate seasons across five years, the predictive capabilities of all models increased, with a maximum R2 of 0.75 during winter. Bidirectional Long Short-Term Memory was the least accurate model for annual training, but had some of the best predictions for seasonal training. Out of five air quality index categories, the XGBoost model was able to predict the correct category 24 h in advance 90% of the time when trained with full-year data. Separated by season, winter is considerably more predictable (97.3%), followed by post-monsoon (92.8%), monsoon (90.3%), and summer (88.9%). These results show the importance of training machine learning methods with season-specific data sets and comparing a large number of methods for specific applications.

show abstract

Section: Discussionmentioning

confidence: 77%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Comparison of Machine Learning Methods to Forecast Tropospheric Ozone Levels in Delhi

Juarez¹,

Petersen

2021

Atmosphere

View full text Add to dashboard Cite

show abstract

“…The growth of renewable energy generation were forecasted and how climate were impacted due to COVID-19 were analyzed using machine learning [7]. Studies carried out to forecast Air Quality using Machine learning [8][9][10][11][12][13][14]. Comparative assessment of ML'S were done to forecast the Troposphere ozone levels [15].…”

Section: Introductionmentioning

confidence: 99%

Repercussions of the COVID-19 Pandemic on the Air Quality of Chennai, India

Vidya¹,

Kumar²,

A³

et al. 2023

Pol. J. Environ. Stud.

View full text Add to dashboard Cite

COVID 19 Pandemic in India had demanded an imposition of lockdown for three weeks initially and was extended further. This has drastic effect on air quality making it better because of control of vehicle emissions. This study analyzed the air quality in Chennai city using the parameters of pollution (NH 3, PM 2.5 , NO 2 , SO 2 , O 3 and CO) for air quality data for monitoring stations (three) spread over the city. National Air Quality Index (NAQI) has been used to present the quality of air spatially during lockdown and before lockdown. The concentrations of PM 2.5 among the pollutants selected showed a maximum reduction (-61%) compared to the pre-lockdown process. NO 2 (−40%) and CO (−32%) have also gone down when lockdown was in place, among other contaminants. In the different sections of the city, about 53% reduction in NAQI has been observed. Deep learning short-term predictions of various air pollutants are made in this study. The correlation between AQI and Pollutants (NH 3, PM 2.5 , NO 2 , SO 2 , O 3 and CO) in the study area modelled in deep learning using PYTHON. The classification of AQI class has been created in python with AQI values of Good (0-50), Satisfactory (51-100), , Very poor (301-400) and severe (>401). The study shows the level of co-relation of PM 2.5 being the highest. A linear regression model was performed and metrics such mean absolute error, r 2 to check the model performance for training and testing data are calculated. These results can be coupled with social, economic and Cultural factors that could have common emission patterns and air quality especially in metropolitan cities. The present study would aid authorities as it clearly shows that the quality can be made better if sources of emission can be diminished. This will pave way to protect and make the surroundings and environment better.

show abstract

Review on air pollution of Delhi zone using machine learning algorithm

Cited by 2 publications

References 28 publications

A Comparison of Machine Learning Methods to Forecast Tropospheric Ozone Levels in Delhi

A Comparison of Machine Learning Methods to Forecast Tropospheric Ozone Levels in Delhi

Repercussions of the COVID-19 Pandemic on the Air Quality of Chennai, India

Contact Info

Product

Resources

About