Agung Bella Putra Utama, Haviluddin, Andri Pranolo, Xiaofeng Zhou, Yingchi Mao
The COVID-19 pandemic had created new challenges for scientists. Forecasting the number of new contaminated cases and deaths is crucial to optimize resource usage and prevent disease transmission. Deep-learning models have improved time-series data processing across different categories. This chapter compares three deep-learning methods, multilayer perceptron (MLP), long-term short memory (LSTM), and convolutional neural network (CNN), to predict the total number of cases and deaths. This analysis forecasted global COVID-19 cases using seasonal and trend data. This chapter used data from 12 countries on daily confirmed cases and fatalities, including the bulk of COVID-19 cases from the World Health Organization (WHO), the USA, India, Brazil, France, Germany, the UK, Italy, Korea, Russia, Turkey, Spain, and Japan. The results reveal that LSTM outperforms other techniques and that deep-learning models can anticipate COVID-19 for seasonal data. © 2024 selection and editorial matter, Aji Prasetya Wibawa.
Universitas Negeri Malang, Malang, Indonesia; Universitas Mulawarman, Samarinda, Indonesia; Hohai University, Nanjing, China; Universitas Ahmad Dahlan, Yogyakarta, Indonesia
