Utilization of Big Data to Predict the Spread of Influenza in the Health System

A major public health issue in Indonesia, influenza is seasonal and seriously harms health as well as the economy. Big data access growth in the healthcare sector opens up fresh opportunities to enhance initiatives for disease prediction and prevention. This paper forecasts the spread of influenza in Indonesia using big data analytics by means of five machine learning algorithms—Random Forest, Long Short-Term Memory (LSTM), Support Vector Machine (SVM), XGBoost, and AdaBoost—that have shown efficacy in handling large-scale and complicated data. While Random Forest was selected for its great accuracy and feature selection ability, LSTM was relied upon to find temporal patterns in time series data. Moreover, SVM was applied because of its ability to classify high-dimensional data; an ensemble learning approach was used to increase model accuracy with XGBoost and AdaBoost. The influenza case data under study came from many sources, including health surveillance systems, meteorological conditions, and population mobility patterns. Many sources provided the influenza case data under study: health surveillance systems, meteorological conditions, and population movement patterns. The five models were applied to determine the main causes of influenza outbreaks and forecast how the illness would spread over various regions of the country. The results show how machine learning and big data can be useful if focused on public health campaigns and make better use of available resources. The health system may reduce the effects of influenza and increase the resilience of Indonesia’s healthcare infrastructure with precise forecasts.