Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal AL-ULUM: JURNAL SAINS DAN TEKNOLOGI
Muhammad Iqbal Firdaus
Fakultas Taknologi Informatika, Universitas Islam Kalimantan Muhammad Arsyad Al Banjari
Agriculture, Biological Sciences & Forestry Chemical Engineering, Chemistry & Bioengineering Computer Science & IT Decision Sciences, Operations Research & Management Energy

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
SELEKSI BEASISWA BIDIK MISI UNISKA MAB BANJARMASIN HIBAH LLDIKTI XI KALIMANTAN MENGGUNAKAN METODE SVM DAN TOPSIS Muhammad Iqbal Firdaus; M Gilvy Langgawan Putra
AL-ULUM: JURNAL SAINS DAN TEKNOLOGI Vol 6, No 1 (2020)
Publisher : Universitas Islam Kalimantan Muhammad Arsyad Al Banjari

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (531.17 KB) | DOI: 10.31602/ajst.v6i1.3655

Abstract

Every year LLDIKTI XI Kalimantan provides scholarships to universities under its auspices. Which Uniska has received scholarships since 2015-2018 as many as 152 bidik misi students for new students. Which is usually selected using manual steps with the help of human power. From the selection process there are problems, namely human factors. Therefore we need a computational process that supports the selection process. Then the SVM (Support Vector Machine) method is used for the classification process and the TOPSIS (Technique For Order Preference By Similarity To Ideal Solution) method is used to give a priority ranking of scholarship. The average speed of the entire process in the selection system and recommendations for the acceptance of the UNISKA Bidik Misi scholarship with the implementation of the SVM and TOPSIS methods using testing from a comparison ratio of 19.12 seconds, the fastest time is 14.40 and the longest time is 23.58. The accuracy of the selection and recommendation of acceptance of the UNISKA Bidik Misi scholarship using the training data comparison ratio and 90%: 10% data testing has an average accuracy of 85.53% and testing based on the best parameters of the SVM sequential training process is λ (Lambda) = 0.1 , constant γ (gamma) = 0.05, ε = 0.0001, Maximum Iteration = 1000, ratio of 90%: 10% and value of d = 2, C (Complexity) = 1. So that the best accuracy is 100% and the average accuracy the best is 93.63%. Keywords: Selection, UNISKA MAB, SVM, TOPSIS, MCDM.
Co-Authors M Gilvy Langgawan Putra