Journal article Open Access

Air Quality Prediction by Classification of Supervised Machine Learning

T. R. Saravanan; V. Pavithra; G.Saranya

Sponsor(s)
Blue Eyes Intelligence Engineering & Sciences Publication(BEIESP)

Generally, air pollution refer to the release of various pollutants into the air which are threatening the human health and planet as well. The air pollution is the major dangerous vicious to the humanity ever faced. It causes major damage to animals, plants etc., if this keeps on continuing, the human being will face serious situations in the upcoming years. The major pollutants are from the transport and industries. So, to prevent this problem major sectors have to predict the air quality from transport and industries .In existing project there are many disadvantages. The project is about estimating the PM2.5 concentration by designing a photograph based method. But photographic method is not alone sufficient to calculate PM2.5 because it contains only one of the concentration of pollutants and it calculates only PM2.5 so there are some missing out of the major pollutants and the information needed for controlling the pollution .So thereby we proposed the machine learning techniques by user interface of GUI application. In this multiple dataset can be combined from the different source to form a generalized dataset and various machine learning algorithms are used to get the results with maximum accuracy. From comparing various machine learning algorithms we can obtain the best accuracy result. Our evaluation gives the comprehensive manual to sensitivity evaluation of model parameters with regard to overall performance in prediction of air high quality pollutants through accuracy calculation. Additionally to discuss and compare the performance of machine learning algorithms from the dataset with evaluation of GUI based user interface air quality prediction by attributes.

Files (750.5 kB)
Name Size
D6782049420.pdf
md5:c05791243b3928d0c8e607cc99de1174
750.5 kB Download
23
9
views
downloads
Views 23
Downloads 9
Data volume 6.8 MB
Unique views 22
Unique downloads 8

Share

Cite as