Article Per Year (5 Year)
p-Index From 2019 - 2024
0.23
P-Index
This Author published in this journals
All Journal JURNAL REKAYASA TEKNOLOGI INFORMASI Jurnal Teknologi Komputer dan Sistem Informasi
Mohammad Guntara
Informatics Department, Universitas Teknologi Digital Indonesia (UTDI), Yogyakarta
Computer Science & IT

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
Drop Out Student Clusterization Using the k-Medoids Algorithm Mohammad Guntara; Totok Suprawoto
JTKSI (Jurnal Teknologi Komputer dan Sistem Informasi) Vol 5, No 1 (2022): JTKSI (Jurnal Teknologi Komputer dan Sistem Informasi)
Publisher : JTKSI (Jurnal Teknologi Komputer dan Sistem Informasi)

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

Student dropout (resign) is a problem that needs to be addressed as early as possible. The number of students dropping out will decrease the quality of the performance of a university, as well as reduce it as much as possible because it will have an impact on public appreciation. As a first step to reducing it, it requires the clustering of students who experience this. Based on this cluster, a pattern of student tendency to drop out can be identified. The parameters used in this study were the GPA, the study period, the number of credits received, and the number of semesters inactive. To compile a cluster, the k-Medoids algorithm is used with 3 types of clusters. Based on the results of the clustering, it can be seen that the dominance of dropout students is due to GPA <2.00 as much as 38.2% and due to not being active in college as much as 52.2%. To measure the cluster quality, the Silhouette coefficient algorithm is used and the resulting coefficient value is 0.3, meaning that the cluster separation rate weak structure.
Analisis Performa Algoritma K-NN Dan C4.5 Pada Klasifikasi Data Penduduk Miskin Femi Dwi Astuti; Mohammad Guntara
Jurnal Rekayasa Teknologi Informasi (JURTI) Vol 2, No 2 (2018): Jurnal Rekayasa Teknologi Informasi (JURTI)
Publisher : Universitas Mulawarman

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (89.285 KB) | DOI: 10.30872/jurti.v2i2.1865

Abstract

Status kemiskinan penduduk di Kecamatan Bantul diklasifikasikan  melalui  11 aspek. Jumlah nilai dari keseluruhan aspek akan menentukan kelas kemiskinan diantaranya kelas miskin, sangat miskin dan rawan miskin. Klasifikasi dengan model tersebut membuat hasil pengelompokan kurang akurat sehingga perlu dicoba klasifikasi dengan model yang lain. Analisis performa klasifikasi data penduduk miskin pada penelitian ini dikerjakan menggunakan metode klasifikasi K-NN dan C4.5. Kedua algoritma klasifikasi akan dibandingkan performanya melalui uji akurasi, precision dan recall.Hasil analisis perbandingan performa algoritma K-NN dengan parameter setting k=1 memiliki performa yang paling baik dibandingkan dengan nilai k=10, 100, 1000 maupun algoritma C4.5. Hasil nilai Accuracy sebesar 94,71%, precision sebesar  84,96% dan recall sebesar 83,6%.
Co-Authors Femi Dwi Astuti Totok Suprawoto