Prediction of Students Graduation Using Decision Tree Method with the Implementation of Algorithm C4.5

Islamic State University of Maulana Malik Ibrahim of Malang is a public college in Indonesia that receives more increased quota of students year to year, but all of students are not able to graduate just in time in accordance with a study period pursued so it results in the accumulation of the number of students who are not graduated in accordance with the graduation periods. Based on that background, the research is conducted to create a system using the technique of classification. The data processing is used to predict unknown-yet class, that is the prediction of students' graduation. Technique of classification used is a decision tree with the implementation of algorithm C4.5. Input used is an attribute of students'data, includingg the origin of the region, types of school, way to university entrance, the experience of pesantren, accumulative grade point average, and grade point average of each semester, from first semester up to fifth semester. That students' data are training sample data used in arranging the decision tree. Based on the testing that use graduated students' data from 2005 up to 2008, accuracy of compatibility in this system reaches 82,79%, so it can be used to predict unknown-yet students' graduation