Air Pollution Prediction in Smart Cities by using Machine Learning Techniques

Abstract

The urban air pollution has an immediate effect on man health specifically in developing and mechanical countries. It can cause health issues such as cancer, cardiovascular diseases and high mortality rates. Continuous checking of contamination empowers the metropolitans to dissect the present traffic circumstance of the city and take their decision accordingly. Existing exploration has utilized diverse AI apparatuses for pollution forecast; notwithstanding, relative examination of these methods is regularly required to have a superior comprehension of their handling time for numerous datasets. In this work, we look at forecasting the air contamination by dealing with parameters of three different gases like SO2,NO2,O3.This process involves to pre-processing the times series. However, pre-processing involves a similarity measure, we explore the use of Dynamic Time Warping (DTW),LSTM,ARIMA Model for time series prediction, K-means, Support Vector Regression is then used to classify the spatio-temporal pollution data of different areas over a period of 10 years.