Google Scholar

Penggunaan Pohon Keputusan Untuk Menganalisis Kemungkinan Pengunduran Diri Calon Mahasiswa Di STMIK AMIKOM Yogyakarta
09 April 2009

Kusrini, Sri Hartati
STMIK AMIKOM Yogyakarta, Universitas Gadjah Mada
e-mail : kusrini@amikom.ac.id, s_hartati@ugm.ac.id

ABSTRACT
In STMIK AMIKOM Yogyakarta, it often happen that a student candidate who has been succeed in the admission test, cancel his/her application by disregarding the next phase of admission process (re-registration). If the withdrawal possibility arised can be detected early, then it is expected that the STMIK AMIKOM executive can make any attempt to keep the candidate proceeds through the admission process and subsequently, minimizes the rate of admission cancellation.
An application to detect the possibility of application withdrawal has been built. The detection is based on the cases occurred in the past. The cases are searched and compared with the current case.
In order to make the cases searching technically easier, an indexing is conducted in the form of a decision tree, and it is used to build the rule bases. The tree is built with C4.5 algorithm that is improvement from the predecessor ID3 algorithm.
More than 1900 data records of new student applicants for 2006/2007 teaching season in STMIK AMIKOM Yogyakarta are used as a set of test case, the experimental results show that the applicants are categorized in registration, not registration and unknown, and result in 508 rules.


Keywords : Decision Tree, C4.5 Algorithm, Aplicant registration.

 

Download >> PENGGUNAAN POHON KEPUTUSAN UNTUK MENGANALISIS KEMUNGKINAN PENGUNDURAN DIRI CALON MAHASISWA DI STMIK AMIKOM YOGYAKARTA.pdf