An Approach to Assess Air Quality using Deep Learning with BRB

Abstract

Air quality estimation is very important to maintain a sustainable world. In this industrial world the environment has adverse effects due to air pollution. Incidents of air pollution is increasing day by day, there is a necessity to predict such occurrence to save human lives. Many expensive sensors have been used to measure this caution; different methods have been applied to solve this problem. Deep learning is a data driven approach which is successfully used to maintain a sustainable environment and also capable to find out hidden features by analyzing enormous data. To assess air quality using deep learning this research used sensor data which may have different kinds of uncertainty. To handle this uncertainty here deep learning technique is integrated with belief Rule based Expert System (BRBES). BRBES is a rule based approach which gives exact prediction based on knowledge base and inference engine. This paper proposed a Convolutional Neural Network (CNN) as a deep learning method which is a combination of convolutional layers and pooling layers to determine multiclass feature, using softmax function. To predict air quality this integrated approach gives remarkable result.