Comparing Clasification Algorithm Of Data Mining to Predict the Graduation Students on Time

by Imam Tahyudin,Ema Utami,Armadyah Amborowati
( Department of Information System STMIK AMIKOM Purwokerto, Department of Magister of Computer Engineering STMIK AMIKOM Yogyakarta )

Date Published: 02 Dec 2013
Published In: Information Systems International Conference (ISICO)
Volume: 2013
Publisher: Departemen Sistem Informasi, Institut Teknologi Sepuluh Nopember
Language: id-ID

Keywords: data mining,classification,Algorithms,Prediction,Student,Graduation,On time

Abstract

The percentage of students who graduate on time is one of the elements of accreditation of a study program. Based on data from the administrative and academic (BAAK) in 2013 showed that the graduation rate of students ontime in STMIK AMIKOM Purwokerto reached 78.80%. As the efforts to increase the percentage of students graduate on time is mining the information from the student database. Trough the student database can classify the levels of the graduation on time. The purpose of this research is to compare the several data mining classification algorithms, especially the Decision Tree (DT), Naive Bayes (NB), Artificial Neural Network (ANN), Support Vector Machine (SVM) and Logistic Regression (LR) algorithms with cross validation evaluation and T -Test to predict the graduation student on time. The method used is the comparison method. Based on the comparison of performance score and t-test, SVM algorithm is the appropriate algorithm that used to predict the student graduation on time. Level of accuracy to predict SVM algorithm is high (almost 100% with excellent clasification category). On the other hand, result of t-test of SVM algorithm is very dominant than other algorithms.


© 2024 Open Access Journal of Information Systems (OAJIS) | created by : radityo p.w (http://about.me/radityopw) and rully a.h (eraha99 [at] gmail.com)