Authors :
K.Sathya; Dr.T.Ranganayaki
Volume/Issue :
Volume 7 - 2022, Issue 8 - August
Google Scholar :
https://bit.ly/3IIfn9N
Scribd :
https://bit.ly/3DqGLKk
DOI :
https://doi.org/10.5281/zenodo.7082027
Abstract :
With increased industry and urbanization, air
pollution is becoming an environmental hazard. Air
Quality (AQ) is becoming increasingly important for
both the environment and humanity. The atmosphere
contaminations that cause air pollution like CO2, NO2,
etc., are produced by the combustion of natural gas, coal
and wood, as well as by industry and cars. Air pollution
may cause serious diseases such as lung cancer, brain
damage, and even death. So, predicting AQ is an
important step for the government to take because it is
becoming a big problem for human health. To predict
AQ, many artificial intelligence frameworks have been
developed over earlier days using the different historical
data on air pollutants in various regions. This
manuscript covers a complete study of various Deep
Learning (DeepLearn) frameworks developed to forecast
AQ using the different available air pollution databases.
First, different AQ prediction frameworks relying on the
DeepLearn structures are discussed briefly. After that, a
comparative study is conducted to understand the
drawbacks of those frameworks and suggest a new
solution to predict the AQ accurately.
Keywords :
Air pollution, Air pollutants, Air quality, Forecasting, Artificial intelligence, DeepLearn.
With increased industry and urbanization, air
pollution is becoming an environmental hazard. Air
Quality (AQ) is becoming increasingly important for
both the environment and humanity. The atmosphere
contaminations that cause air pollution like CO2, NO2,
etc., are produced by the combustion of natural gas, coal
and wood, as well as by industry and cars. Air pollution
may cause serious diseases such as lung cancer, brain
damage, and even death. So, predicting AQ is an
important step for the government to take because it is
becoming a big problem for human health. To predict
AQ, many artificial intelligence frameworks have been
developed over earlier days using the different historical
data on air pollutants in various regions. This
manuscript covers a complete study of various Deep
Learning (DeepLearn) frameworks developed to forecast
AQ using the different available air pollution databases.
First, different AQ prediction frameworks relying on the
DeepLearn structures are discussed briefly. After that, a
comparative study is conducted to understand the
drawbacks of those frameworks and suggest a new
solution to predict the AQ accurately.
Keywords :
Air pollution, Air pollutants, Air quality, Forecasting, Artificial intelligence, DeepLearn.
