Jurnal Ilmiah Sinus
Vol 17, No 1 (2019): Vol. 17 No. 1 Januari 2019

OPTIMASI METODE NAIVE BAYES DENGAN FEATURE SELECTION INFORMATION GAIN UNTUK PREDIKSI KETERLAMBATAN PEMBAYARAN SPP SEKOLAH

Muqorobin Muqorobin (Program Pascasarjana Teknik Informatika, Universitas Amikom Yogyakarta)
Kusrini Kusrini (Program Pascasarjana, Universitas Amikom Yogyakarta)
Emha Taufiq Luthfi (Program Pascasarjana, Universitas Amikom Yogyakarta)



Article Info

Publish Date
29 Jan 2019

Abstract

The cost of education is one component of input that is very important in implementing education. Because costs are the main requirement in an effort to achieve educational goals. SMK Al-Islam Surakarta is a private education institution that requires students to pay school fees in the form of Education Development Donations. Educational Development Donation is a routine school fee that is conducted every month. Based on last year's TU report, many students were late in paying Education Development Donations, around 60%. This is a big problem. The purpose of this study is that researchers will build a predictive system using the Naïve Bayes method. Because the method can classify the class right or late, in the payment of school fees. Data processing was taken from the dapodik data of schools in 2017/2018 with the test dataset taking 30 records. To find out the level of accuracy, this research was conducted with the Naive Bayes Method and the Information Gain Method for feature selection. Accuracy testing is done by the Confusion Matrix method. The results showed that the highest accuracy was obtained by combining the Naive Bayes Method with the Information Gain Method obtained by 90% accuracy. 

Copyrights © 2019






Journal Info

Abbrev

e-jurnal_SINUS

Publisher

Subject

Computer Science & IT

Description

Jurnal Ilmiah SINUS is a magazine published twice a year, wherein one issue there are seven articles. Jurnal Ilmiah SINUS as a communication medium to report the results of field research, library research, observations or opinions on problems arising related to the development of information ...