Improved Malaria Outbreak Predictive Model Using Naïve Baye and Artificial Neural Network

Malaria is one of the deadliest diseases in West Africa sub-region that needed urgent address, especially during the rainy season. Early alert of the disease outbreak can come a long way in degrading its devastating effect on the community, and also help decision-makers understand the gravity of the disease in our community as well as guide them in providing pro-active solutions. This study developed an improved malaria outbreak model using Naïve Bayes and Artificial Neural Network using a large dataset from Kebbi state, Nigeria, for the year 2020-to-2022 considering parameters such as; Min. Temperature, Max. Temperature, Humidity, Number of Rainfall, and Number of cases recorded, and outbreak recorded in YES/NO. The study checked accuracy using a confusion matrix and finds that Naïve Bayes predict better with high accuracy of 90% than Artificial Neural Network at 98%.