The Empirical comparison of decision tree approaches with classification task for finding graduate employment status

Abstract

Classification task is one of the most useful techniques in data mining, which is to
find out meaningful and useful pattern in large volumes of data. For classification task,
Decision tree approaches are the most commonly used because of its facility of
implementation and easy to understand. The study purpose various approaches available for
classification like J48, Simple Cart, Random Forest, LAD Tree, REP Tree, Decision Stump
etc. The research we introduce eight algorithms from them. We take graduates dataset in
academic year of 2011 Maejo University Thailand. Implement theses algorithms in graduate
dataset and compare TP-rate, Fp-rate, Precision, Recall and ROC Curve parameter.