Air Pollution Forecasting and Performance Evaluation Using Advanced Time Series and Deep Learning Approach for Gurgaon

Singh, Ankit (2019) Air Pollution Forecasting and Performance Evaluation Using Advanced Time Series and Deep Learning Approach for Gurgaon. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration manual) Download (1MB) | Preview

Abstract

Due to its detrimental repercussions, air pollution has been an significant research area since last few years. Pollution forecasting demands advanced monitoring stations along with complex algorithms to evaluate time related pollutant data. Accurate air quality forecasting is critical for systematic pollution control as well as public health and wellness. An Indian city Gurugram has been ranked as the highest polluted city in the world since last two years by AirVisual. Advanced forecasting models can be implemented on considerable pollutant datasets to get valuable future forecasts that aid government health organisations to adopt precautionary measures. This study involved the utilisation of novel forecasting model Prophet to predict the future pollution precisely. The obtained results were compared with several statistical, time series and deep learning models such as AR, ARMA, ARIMA, SARIMA, Exponential Smoothing, TBATS and LSTM in terms of forecasting error and other factors. Model evaluation has been carried out on 8 hourly pollution data using multiple evaluation metrics such as RMSE, MSE, MAE and MAPE. The results obtained have been evaluated and visualised. Research findings signify Prophet to be the most efficient forecasting model with the lowest combined evaluation errors including RMSE, MAE, MSE and MAPE proving capable of handling outliers, trend and seasonality in data. Prophet also gave good performance on a separate Delhi data. This approach can assist in improving the current forecasting quality thereby benefiting the country and its people.
Index Terms : Air Pollution, Forecasting, AQI, Time Series, Deep Learning, PROPHET

Item Type:	Thesis (Masters)
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > QA Mathematics > Computer software T Technology > T Technology (General) > Information Technology > Computer software G Geography. Anthropology. Recreation > GE Environmental Sciences > Environment
Divisions:	School of Computing > Master of Science in Data Analytics
Depositing User:	Dan English
Date Deposited:	10 Jun 2020 16:51
Last Modified:	10 Jun 2020 16:51
URI:	https://norma.ncirl.ie/id/eprint/4266

Actions (login required)

View Item