PEMANFAATAN METODE KLASIFIKASI UNTUK PREDIKSI KEBERHASILAN SISWA SMA DITERIMA PERGURUAN TINGGI (Studi Kasus : SMA N 1 Ponorogo); USING CLASIFICATION METHOD FOR PREDICTING SENIOR HIGH SCHOOL STUDENT ENTRANCE TO HIGHER EDUCATION (Case Study : SMA N 1 Ponorogo)

Muhammad, Agus Salman, Edi Winarko

2015 | Disertasi | FMIPA

Now a days the amount of data stored in educational database increasing rapidly. These database contain hidden information for improving of student’s performance. Educational data mining is used to study the data available in the educational field and bring out the hidden knowledge from it. Classification methods like decision tree, Bayesian network etc can be applied on the educational data for predicting the student’s performance in examination. This prediction will help to identify student accept in the university and help them to score better marks. The C4.5 , Naive Bayes and Support Vektor Machine algorithm are applied on senior high school student’s to predict their performance. The outcome of the C4.5, Naive Bayes, Support Vektor Machine predicted the degree who likely to pass in the university. The comparative analysis use data catur wulan, semester result and numerik format. To analyse the accuracy of the C4.5, Naive Bayes, Support Vektor Machine algorithm use semester result. The higher result accuracy prediction use algorithm C4.5. The accuracy of algorithm C4.5 is 85.316 %, Naive Bayess methode is 48.354 % and Support Vektor Machine is 69.842 %. Correlation based Feature Selected (CfBS) methode use to selection much attribute to any attribute. The result of greedy stepwise with algoritma C4.5 is 81.868 %. The differences between all attribute and selected attribute is 3.448 %.

Kata Kunci : Education data mining; clasification, prediction; fitur selection; catur wulan result; semester result.