Penentuan Tingkat Keberhasilan Mahasiswa Tingkat I IPB Menggunakan Induksi Pohon Keputusan dan Bayesian Classifier

Abstract

Tingkat Persiapan Bersama (TPB) or Collective’s Preparation Level is the term that is used to call the first-year bachelor degree students in IPB (Bogor Agricultural University). IPB decides these first-year students graduation from their academic result in the end of TPB year. The students can continue to the next education year if they complete many graduation requirements that was decided before. The success of the first year IPB’s students can be looked from their academic graduation result. One of techniques can be used to determine student success is data mining. Data mining is used to build classifier that shows the success level of the first year student of IPB. This research aimed to develop classifier to describe the level of IPB’s students success and predict the new student of IPB. This research uses two data mining method, that is decision tree and Bayesian classifier (naïve Bayes). Decision tree is used to describe the level of IPB’s students success and to get crusial factor that determine IPB’s students success in their first year. Naïve Bayes is used to predict the new student of IPB. The results this research are a probabilistic classifier naïve bayes with accuracy 57,160 % and a decision tree classifer that contains 3 classifiaction rules with accuracy 63,542%. According to the rules that we got from decision tree, the factor that influence the first year IPB student 2007/2008 is point of high school final examination.