Hybrid Intelligent Modelling for Air Quality Prediction Deep Learning and Markov Chain Unconventional Framework
Roba Zayed and Maysam F Abbod (Brunel University London, United Kingdom (Great Britain))
The purpose of this research is to build an innovative prediction model, define measurable (quantifiable) data and use them to measure air quality in selected cities. This study presents a multivariate hybrid Markov-switching dynamic model using a multi-state transition method for multiple outputs and a Deep Neural Network through a niche experimental framework. The experiment is part of applied big-data AI research which aims to predict air quality and present a reliable system which will provide an air-quality index using hybrid model. This will become a tool for decision-makers concerned with related air-quality issues. This research presents a multi-input multi-output hybrid model with reliable accuracy of hourly time-series data, and provides the large dataset in this study. This aims to cover the gap in high big-data prediction accuracy for the domain (hourly frequency) and to form a more standardized air-quality index by comparing results in two selected cities: London and Jordan.
Journal: International Journal of Simulation- Systems, Science and Technology- IJSSST V23
Published: Apr 1, 2022