Developing spatiotemporal prediction models of the DF transmission using Cellular Automata (CA), ARIMA, RNN, LSTM, CNN, and TCN -- 2

I'm doing a research project to design several prediction modelling and compare them to achieve the best; I have numerous methods that I want to apply to conduct my analysis. The analysis required experts aware of using ArcMap/QGIS, Machine learning, and deep learning methods to achieve the project objectives. The Project needs to be done fast and in high quality. So, I need to develop spatiotemporal models that can get implemented for how the effective disease is going to spread over time and space "not classification". it can be done based on the location and the number of affected people each week, month and year. The required models are RNN, CNN, LSTM, TCN (Temporal Convolutional Network), GRU and a combination of this algorithm and cellular automata. in short: Through this project I expect: - We want to know how many people are expected to be affected next week, next month, and next year in addition to the risk areas. - What is the link between disease transmission and other factors (weather, elevation, land uses land cover) - The used models are ( RNN, CNN, LSTM, TCN (Temporal Convolutional Network), GRU, and cellular automata). Then compare the best model among these. - Apply it in ArcMap/QGIS to get spatiotemporal visualization for the risk areas.