AAEDM: Theoretical Dynamic Epidemic Diffusion Model and Covid-19 Korea Pandemic Cases

Kim, Song-Kyoo

This paper deals with an advanced analytical epidemic diffusion model which is capable to predict the status of epidemic impacts. This newly propose model well describes an epidemic growth and it could be widely applied into various topics including pathology, epidemiology, business and data sciences. The Advanced Analytical Epidemic Diffusion Model (AAEDM) is a dynamic diffusion prediction model which is theoretically intuitive and its tractable closed formula could be easily adapted into versatile Bigdata driven analytics including the machine learning system. This dynamic model is still an analytical model but the periods of prediction are segmented for adapting the values from the dataset when the data is available. The epidemiologically vital parameters which effect on the AAEDM are also introduced in this paper. The evaluation of this theoretical model based on the Covid-19 data in Korea has been accomplished with relative fair future prediction accuracies. Although this analytical model has been designed from a basic exponential growth model, the performance of the AAEDM is competitive with other Bigdata based simulation models. Since the AAEDM is relatively simple and handy, anyone can use this model into analyzing outbreak situations in his daily life.