Decision Tree C 4.5 Algorithm for Classification of Poor Family Scholarship Recipients

Abstract

Abstract The poor family scholarship program is a scholarship program by Lamongan District Education Office. This scholarship is given to underprivileged students for study costs at university. The many indicators of scholarship determination result in difficulties education office in making objective decisions. The purpose is to determine the classification of scholarship recipients using the decision tree C4.5 algorithm. This algorithm has the advantage of speed informing and reading models, high accuracy, flexibility, and processing discrete and continuous data. The contribution of this research is to classify poor family scholarship participants with the C 4.5 algorithm without pruning, pre pruning, and post pruning methods. This study uses 460 data and for each attribute, it is calculated using the gain ratio formula to produce trees and decision rules. Based on the scenario that has been done the accuracy value generated after pruning is better than without pruning. The classification process using scenario post pruning technique is used as a reference for the classification of new registrants because it has the highest accuracy of 85.55%.