Muhammad Haviz Irfani
Universitas Indo Global Mandiri, Palembang

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

Found 1 Documents
Search

Prediksi Calon Mahasiswa Penerima KIP Pada Universitas Indo Global Mandiri menggunakan Algoritma Decision Tree Zakaria Saputra; Dewi Sartika; Muhammad Haviz Irfani
Resolusi : Rekayasa Teknik Informatika dan Informasi Vol. 4 No. 3 (2024): RESOLUSI January 2024
Publisher : STMIK Budi Darma

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30865/resolusi.v4i3.1501

Abstract

Indo Global Mandiri University has accepted Smart Indonesia Card students since 2020. Financing for Smart Indonesia Card students is obtained from government subsidies on the condition that students are required to complete lectures for less than or equal to four years and maintain academic achievement with a minimum cumulative grade point average of 3.00. As in the Informatics Engineering study program, 31 prospective Smart Indonesia Card recipient students must be selected first to fill the available quota. Students who have passed the selection will be ranked to determine the number of prospective students who are entitled to receive assistance carried out by the university's Cooperation Bureau. The purpose of the research is to do programming that can predict the number of students receiving this assistance using the Decision Tree algorithm. Decision Tree (C4.5 version) is one of the classification methods that uses a tree structure representation where each node represents an attribute, the branch represents the value of the attribute, and the leaf represents the class. The attributes used are skill program, average score, KK status, income, and aid card. Based on calculations carried out using the decision tree method, the value of the percentage of accuracy is quite good with a value of 73% for predicting prospective KIP recipient students at Indo Global Mandiri University Palembang, while the performance of the classification model still has a fairly low performance with a precision value of 40%, a recall value of 56% and an F1-score value of 47%.