Real Time Air Pollution Prediction in Urban Cities using Deep Learning Algorithms and IoT

Air pollution is a major cause in many urban cities. Particulate Matter (PM) present in air cause many health issues like lung cancer, asthma, cardiopulmonary disease, and environmental changes. Using the Internet of Things (IoT) to bring environmental protection into the 21st century, this research study offers a real-time air pollution monitoring and forecasting system that may be used to monitor and predict air pollution in real time. A deep learning approach called as Long Short-Term Memory (LSTM) with ADAM optimization may be used to estimate the amount of particulate matter (PM) in the air. This LSTM framework can be used to predict pollution particles before an hour. The collected sensor data are based on time series, for which LSTM model is well suited. ADAM optimization is well suited for learning data and analyse it deeply. It also exhibits cutting edge of Artificial Intelligence (AI), instead of Machine Learning (teaching of computers to process and learn from data). With Deep Learning (DL) the computer is trained itself to learn and process data. This paper proposes executing a huge air pollution dataset in Keras framework using SPYDER tool with python. Pollution in a certain location may be anticipated an hour in advance using this technology, and traffic can be diverted to a different route.