Measles Prediction Modeling using Adaptive Neuro-Fuzzy Inference System (ANFIS) and Artificial Neural Network (ANN)

Measles, a viral and communicable illness with life-threatening complications, begins with an 8-12-day incubation period accompanied by fever, cough, coryza, and conjunctivitis. The Measles Virus (MeV) enters the alveolar macrophages of the host or its dendritic cells, replicates, and then spreads to the surrounding surfaces. As a result, we focus on the use of Adaptive Neuro-Fuzzy Inference Systems (ANFIS) and Artificial Neural Networks (ANN) for the prediction of measles. The models (ANFIS and ANN) were employed and evaluated using the coefficient of determination (R2), Root Mean Square Error (RMSE), Pearson correlation coefficient (R), and Percentage Bias (PBIAS) as evaluation metrics. The results showed that ANFIS, ANN, and MLR showed R values of 0.9810, 0.9960, and 0.9750 for both training and test. Remarkably, ANN scored lower in terms of the RMSE metric with values of 0.6790 and 0.6809 for both training and testing. In addition, the ANN PBIAS values were 0.1080 and 0.1142 for both training and testing, respectively. In conclusion, ANFIS and ANN have demonstrated effectiveness in modeling measles occurrences, thus, as a valuable tool for public health and decision makers, especially in determining strategies that will enhance public safety.