Determining the factors affecting the death of children aged in Turkey and modeling with artificial intelligence methods

Deaths under the age of five are divided into infant and childhood deaths. Deaths occurring at or before the 11th month are called infancy deaths, and deaths occurring between the 12th and 59th months are called childhood deaths. Countries are carrying out many different studies to reduce the deaths of children under five years old. With the studies carried out, deaths are decreasing, but the targeted results have not yet been achieved. In this study, the effects of the factors affecting the deaths of children under the age of five between 1988 and 2018 on death outcomes are examined. In the study, 10 features are evaluated with 9 different algorithms using the Orange 3 28.0 program. It is aimed to guide the reduction of deaths of children under the age of five by identifying the attributes that have the most impact on the deaths of children under five years old. In order to increase the classification success of the algorithms used in the study, hyperparameter optimization was used. In this way, the parameters that give the best results of the algorithms can be determined. According to the results obtained by using algorithms and parameters suitable for the study, the most effective attribute on the deaths of children under the age of five was gross domestic product per capita. The most successful result is achieved with the Support Vector Machines algorithm.