Forecasting Measles in the European Union Using the Adaptive Neuro-Fuzzy Inference System

BACKGROUND/AIMS Measles is one of the diseases that cause child mortality. The measles forecasting is essential in planning the fight against the disease and reducing the risk of the vaccine stocks expiration. Governments and health institutions estimate the measles vaccine requirements using certain equations, which are generally based on the size of the target population and the past consumption records. There ore several studies that have examined the measles forecasting and conducted a vaccine requirement assessment. MATERIAL and METHODS This study uses a forecasting model that employs an adaptive neuro-fuzzy inference system (ANFIS) based on clustering. In this study, the measles data were derived using the World Health Organization (WHO) Measles and Rubella Surveillance Data, which cover the period from January 2011 to March 2018 and include 28 European Union member countries. Out of total 87 monthly measles cases, 80% were used for training, and 20% were chosen for testing. RESULTS In addition to the mean square error, the root mean square error, normalized root mean square error, mean absolute error, and mean absolute percentage error were calculated. CONCLUSION The model created for this purpose has demonstrated that the predictions made for the data collected between January 2011 and March 2018 were successful.