Predicting the Spread of the Corona Virus (COVID-19) in Indonesia: Approach Visual Data Analysis and Prophet Forecasting

Abstract

The development trend of the coronavirus pandemic (COVID-19) in various countries has become a global threat, including in Southeast Asia, such as Indonesia, the Philippines, Brunei, Malaysia, and Singapore. In this paper, we propose an Exploratory Data Analysis (EDA) model approach and a time series forecasting model using the Prophet method to predict the number of confirmed cases and cases of death in Indonesia in the next thirty days. We apply the EDA model to visualize and provide an understanding of this pandemic outbreak in various countries, especially in Indonesia. We present the trends in the spread of epidemics from the countries of China from which the virus originates, then mark the top ten countries and their development and also present the trends in Asian countries. We present an analytical framework comparing the predicted results with the actual data evaluated using the MAPE and MAE models, where the prophet algorithm produces good performance based on the evaluation results, the relative error rate of our estimate (MAPE) is around 6.52%, and the model average false 52.7% (MAE) for confirmed cases, while case mortality was 1.3% for the MAPE and MAE models around 236.6%. The results of the analysis can be used as a reference for the Indonesian government in making decisions to prevent its spread in order to avoid an increase in the number of deaths

Peramalan Deret Waktu untuk Bisnis : Pendekatan algoritma LGBM Regressor

“…Penelitian ini melaporkan bahwa pendekatan regresi untuk peramalan penjualan dapat memberikan hasil yang lebih baik dibandingkan dengan metode deret waktu. Selain itu pendekatan algoritma LightGBM [17], [18] maupun Prophet [19] merupakan salah algoritma yang perlu di pertimbangkan dalam membangun peramalan yang optimal dalam bisnis [20]- [23].…”

Lessons learned from three Southeast Asian countries during the COVID-19 pandemic

“…Several studies have investigated issues such as the impacts of the pandemic on Malaysian tourism ( Karim, Haque, Anis, & Ulfy, 2020 ), clinical analysis ( Sim et al, 2020 ), mathematical models ( Alsayed, Sadir, Kamil, & Sari, 2020 ; Mohd & Sulayman, 2020 ; Salim et al, 2020 ; Yusoff & Izhan, 2020 ), and the society preparedness ( Azlan, Hamzah, Sern, Ayub, & Mohamad, 2020 ; Shah et al, 2020 ). For Indonesian cases, there are limited studies (e.g., Djalante et al, 2020 ; Husein, Hutabarat, Sitorus, Giawa, & Harahap, 2020 ) and only one study ( Husein et al, 2020 ) discussed the epidemic model. That study ( Husein et al, 2020 ) also did not analyse the impacts of lockdowns and social behavior in the COVID-19 transmission.…”

“…For Indonesian cases, there are limited studies (e.g., Djalante et al, 2020 ; Husein, Hutabarat, Sitorus, Giawa, & Harahap, 2020 ) and only one study ( Husein et al, 2020 ) discussed the epidemic model. That study ( Husein et al, 2020 ) also did not analyse the impacts of lockdowns and social behavior in the COVID-19 transmission.…”

“…Moreover, there are very limited studies providing mechanistic models to reproduce the COVID-19 in Southeast Asia, especially in Malaysia and Indonesia. Those studies (e.g., Salim et al, 2020 ; Husein et al, 2020 ), for instance, discussed epidemic models based on statistical fitting which is not useful as statistical fitting models aim to fit models with observed data, lacking capability to conduct policy analysis ( Holmdahl & Buckee, 2020 ).…”

Sistem Peringatan Dini Untuk Pengendalian Pembatasan Jarak Fisik Dengan Metode RSSI Menggunakan Modul Wemos D1 Mini

“…17, No. 4, Desember 2021 [5]. Pembatasan ini dalam penerapannya meliputi proses belajar mengajar dalam jaringan (daring), bekerja dari rumah (work from home -WFH), melarang kegiatan yang mengumpulkan orang banyak, dan pembatasan jam operasi di tempat umum seperti stasiun dan halte bus.…”