Machine learning classification of infectious disease distribution status

Abstract

Infectious diseases are common diseases and are caused by microorganisms such as viruses, bacteria, and parasites. Indicators of the spread of this disease can be seen based on the population level and the number of confirmed cases. This study aims to develop a machine learning (ML) analysis model using the K-means cluster, artificial neural network (ANN), and decision tree (DT) methods. The dataset used in this study was obtained based on the number of confirmed patients and the distribution of the population. The analysis process is divided into two stages, namely preprocessing and the classification process. The pre-processing stage aims to produce a classification pattern that can describe the level of distribution status. The classification pattern will be continued at the classification analysis stage using ANN and DT. Classification analysis gave significant results with an accuracy rate of 99.77%. The results of the classification analysis can also describe the level of knowledge distribution based on the decision tree. Overall, the contribution of this research is to develop a classification analysis model that presents the latest information and knowledge. The results of the research presented also have an impact on the control process in environmental management and public health.