Prediksi Tingkat Kelulusan Mahasiswa Menggunakan Machine Learning dengan Teknik Deep Learning

Martanto Martanto, Irfan Ali, Mulyawan Mulyawan

Abstract


The graduation rate of students on time at the Informatics Engineering study program STMIK IKMI Cirebon greatly affects the accreditation assessment. Graduation prediction is difficult to do, but many have done predictions using a variety of methods. Graduation prediction is needed in order to determine preventive policies for students who graduate not on time. The method used in this research is Machine learning with deep learning techniques. The data set used as many as 405 data of students who graduated on time or who were not on time. The research attributes used are the Nim attribute, the GPA value of students who have graduated and the status of graduating or not graduating. The results of this study are the level of accuracy using Machine Learning by 72.84%.

Keywords


Prediction; Graduation; Machine Learning; Deep Learning

Full Text:



DOI: http://dx.doi.org/10.30591/jpit.v4i2-2.1877

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.

Terindeks oleh :

 

 http://ejournal.poltektegal.ac.id/public/site/images/informatika/Google_Scholar_logo.png

 

 

 

 

 

 

 

   ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Tim Redaksi JURNAL INFORMATIKA : JURNAL PENGEMBANGAN IT

Program Studi D4 Teknik Informatika
Politeknik Harapan Bersama Tegal
Jl. Mataram No.09 Pesurungan Lor Kota Tegal

Telp. +62283 - 352000

Email :
informatika.ejournal@poltektegal.ac.id

   

Copyright: JPIT (Jurnal Informatika: Jurnal Pengembangan IT) p-ISSN: 2477-5126 (print), e-ISSN 2548-9356 (online) 

Flag Counter
 
 
 
 
site
stats
 
View Visitor Statistic
 
 
 
 
 

 

Creative Commons License
JPIT (Jurnal Informatika: Jurnal Pengembangan IT) is licensed under a Creative Commons Attribution 4.0 International License.