A modified decision tree algorithm c4.5 for autism children classification problem

Abstract

This research is aimed at improving the techniques used to manipulate the group to build support systems in predicting autism in children. From basic knowledge of autism in children, the author uses a genetic algorithm to optimize the results of C4.5 decision tree and launches autism diagnosis process. In addition, the author changes the symptoms of the disease into attributes of the input data and the conclusion of disease into attributes of the output data and then proceed to install the application. This research also contributes to developing methodologies to serve in predicting autism in children, thereby helps parents, teachers and doctors detect the disease early in order to improve efficiency in the treatment.