Smart System for Dengue Fever Diagnosis: A Machine Learning Approach

Abstract

Dengue fever is a serious illness that can lead to death in areas where epidemics spread and in third world countries. Early diagnosis is crucial in preventing the severity of the disease and avoiding fatalities. To address this issue, a smart Android application has been developed that uses machine learning algorithms such as the decision tree to diagnose dengue patients. The decision tree algorithm was found to be the most accurate, with an accuracy rate of 93.7%, while other algorithms like close neighborhood had an accuracy rate of 74.29%, na�ve bays had an accuracy rate of 93.07%, and SVM had an accuracy rate of 68.32%. The system's response is based on specific data that is entered and processed through the decision tree algorithm. Overall, the development of this smart system can greatly improve early diagnosis of dengue fever and potentially save lives.