An Intelligent Air Pollution Prediction System Using Deep Learning Techniques

Abstract

Smart city needs to ensure the quality air without any pollution. Recently, the air pollution is increasing all the urban areas including the top metropolitan cities of various developed and developing countries in the world. Air pollution level is increasing frequently due to the development of urban areas, traffic, more number of vehicles use, not aware about the air pollution and factories. This air pollution leads to bring various diseases including cardio vascular disease, asthma, other breathing related diseases for the human being irrespective of age and gender. The smart cities of Malaysia include Kuala Lumpur and Johor Bahru also severely affected due to the traffic congestion. This is because, an effective air quality prediction system is proposed in this work to predict the air quality and the pollution level by using various data preprocessing algorithms and classifiers. newlineThe systemic way of air pollution prediction process is done in world-wide by using Machine Learning (ML) and Deep Learning (DL) for measuring the air pollution level of the country. Many air quality prediction system and air pollution prediction systems are developed by various researchers in the past and achieved better performance in terms of prediction accuracy and error rates. Generally, the smart city concept is created by combining the Information and Communication technology and fixed or mobile sensors. These are all the sensors are fixed in the cities and observe and handle the devices and human beings as well in the city. In this scenario, a new intelligent air quality prediction system is necessary to predict the air quality as much as possible and the relevant instructions needs to be passed to the concern without any delay and it also used to make a decision about the precaution action against the air pollution. newlineTo address the issues of air quality prediction systems and air pollution prediction systems, a new intelligent air quality prediction system is developed in this research by using many data pre-processing and c