JURNAL MEDIA INFORMATIKA BUDIDARMA
Vol 7, No 4 (2023): Oktober 2023

Metode MICE Support Vector Machine (MICE-SVM) untuk Klasifikasi Performance Mahasiswa Merdeka Belajar Kampus Merdeka

Angga Apriano Hermawan (Universitas Dian Nuswantoro, Semarang)
Galuh Wilujeng Saraswati (Universitas Dian Nuswantoro, Semarang)
Etika Kartikadarma (Universitas Dian Nuswantoro, Semarang)



Article Info

Publish Date
22 Oct 2023

Abstract

The Ministry of Education and Culture established a Merdeka Belajar Kampus Merdeka (MBKM) program with the aim of improving the competency of student graduates, both soft skills and hard skills, so that they are better prepared and relevant to the needs of the times, preparing graduates as future leaders of the nation who are superior and have personality. However, the MBKM program is not always effective in improving the quality of a student because there are still several shortcomings. It is also felt that some students have not received maximum results when participating in the MBKM program. In fact, not all programs offered by MBKM partners receive an assessment in the form of soft skills scores. The aim of this research is to classify whether the MBKM program influences the performance of MBKM program students by applying the Multivariate Imputation by Chained Equation (MICE) method to overcome missing values in the classification of MBKM student performance at the Faculty of Computer Science, Dian Nuswantoro University. The qualification of MBKM student performance is very important because we need to know whether the program is deemed effective or not to be continued in the future. In this study, researchers used a dataset originating from the MBKM report from students at the Faculty of Computer Science, Dian Nuswantoro University. Researchers obtained data by collecting data from MBKM student certificates and reporting the results. The data taken was 277 pieces for training and 69 pieces for testing. Next, the researchers used the Support Vector Machine (SVM) algorithm for the classification process. The research results show that the performance of the Support Vector Machine (SVM) algorithm model with MICE missing value handling has better accuracy results, with an accuracy value of 98.07% compared to using the Mean Imputation method, which only obtains an accuracy of 97.34%.

Copyrights © 2023






Journal Info

Abbrev

mib

Publisher

Subject

Computer Science & IT Control & Systems Engineering Electrical & Electronics Engineering

Description

Decission Support System, Expert System, Informatics tecnique, Information System, Cryptography, Networking, Security, Computer Science, Image Processing, Artificial Inteligence, Steganography etc (related to informatics and computer ...