Prediction of Dengue using Machine Learning Algorithms: Case Study Dhaka

In recent years, Bangladesh and the South Asian countries are facing a huge problem from mosquito-borne diseases like dengue fever. According to WHO, the global dengue incidence has grown dramatically in recent decades and there are an estimated 100–400 million infections each year. Approximately 100 million people get sick from infection, and 22,000 die from severe dengue. If we can predict the high time of dengue with environmental/meteorological factors then we can take precautionary steps to mitigate the dengue outbreak. The proposed prediction model can accurately predict this epidemic with machine learning algorithms. In this research, environmental factors like rainfall, humidity, temperature of Dhaka city are considered. The proposed machine learning model uses Support Vector Machine (SVM) for prediction which performs better than the other models used for comparison. Precision, recall, f1 score, accuracy score are used as the measurement of the performance. We got 97% accuracy in SVM for predicting the dengue epidemic in Bangladesh. Most of the research works in this area have been focused on visualizing the dengue scenario whereas this paper contributes an efficient machine learning model with limited available data for predicting the dengue.