EPIDEMIC PROGNOSIS: COMPARATIVE PERFORMANCE OF MACHINE LEARNING AND DEEP LEARNING MODELS FOR PREDICTING VIRUS TRANSMISSION DYNAMICS

The transmission of viral diseases, such as COVID-19, influenza, and other viral strains, poses a substantial worldwide challenge. In the context of health, it is necessary to possess a comprehensive comprehension, meticulous examination, and precise anticipation of the dissemination of this infectious disease. Nonetheless, the presence of diverse data characteristics among different nations poses a considerable obstacle in the development of prediction models for assessing the transmission, mortality, and recovery rates in Indonesia. Understanding the intricacies of viral transmission poses a significant hurdle because to the fluctuating nature of the generalization rate, which is contingent upon country-specific data.The research entailed a comparison of different predictive models, including Random Forest, Simple Linear Regression (SLR), Gaussian Naive Bayes, Multi-Layer Perceptron (MLP), H2O, and Long Short-Term Memory (LSTM), with the purpose of predicting viral transmission. The evaluation metrics encompass MAE, RMSE, and MAPE. The outcomes of the examination of comparison models will aid in identifying the most suitable model for forecasting the transmission of the virus, encompassing the rates of recovery, death, and positive cases, within the specific setting of Indonesia. This work has significance in elucidating the inherent trade-off between efficiency and accuracy within the realm of dynamic data modeling, specifically in the context of COVID-19 viral data.