Predicting degree-completion time with data mining

Data mining in academic databases nowadays used for analyzing patterns and gaining new useful knowledge. This paper tries to predict the degree-completion time of bachelor's degree students using data mining technique and algorithms especially C4.5 and naive Bayes classifier algorithm, and measure the algorithms accuracy, precision, and recall percentages for both algorithms also exploring some factors that assume in theory have some impact on the model. The result from given dataset to build the models shows that C4.5 algorithm better than naive Bayes classifier algorithm with 78% accuracy, 85% weighted mean class precision, and 65% weighted mean class recall. This research can be expanded with different data mining algorithms or other related attributes that have some effects to the degree-completion time.