Study of Climate Factors Affecting the Incidence Rate of Dengue Fever Using Long Short-Term Memory (LSTM) for Mixed Data

This study aims to investigate the meteorological factors affecting the incidence of dengue fever in Thailand using machine learning, specifically employing a Long Short-Term Memory (LSTM) neural network model. The data utilized is a mixed data, collecting from various provinces. for 10 years (January 1, 2013to December 31, 2022), including monthly dengue fever cases from the disease surveillance system (report 506) of the Department of Disease Control, Ministry of Public Health, and monthly meteorological data from NASA/POWER. The dataset was divided into 80% for training and 20% for testing. It has been found that the LSTM neural network model is efficient in studying the meteorological factors influencing the incidence of dengue fever, with R2 value of 0.893 and RMSE value of 28.019. In addition, the meteorological factors significantly affecting the incidence of dengue fever are as follow Total rainfall amount at 2 meters, Wind direction at 2 meters, Surface temperature of the Earth, Maximum wind speed at 2 meters, Relative humidity at 2 meters, Specific humidity at 2 meters, Wind speed at 2 meters, Maximum surface temperature at 2 meters, Average rainfall amount at 2 meters, Surface temperature at 2 meters, Maximum wind speed at 2 meters, Maximum surface temperature at 2 meters and Surface pressure, respectively.