Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Vergy Ayu Kusumadewi
Fakultas Ilmu Komputer, Universitas Brawijaya
Computer Science & IT Control & Systems Engineering Education Electrical & Electronics Engineering Engineering

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

Found 1 Documents
Search

 1 
Klasifikasi Jurusan Siswa menggunakan K-Nearest Neighbor dan Optimasi dengan Algoritme Genetika (Studi Kasus: SMAN 1 Wringinanom Gresik) Vergy Ayu Kusumadewi; Imam Cholissodin; Putra Pandu Adikara
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 4 No 4 (2020): April 2020
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (774.76 KB)

Abstract

Majors is the process of selecting and placing study programs that are suitable for students, this process will affect the future of students, both when they become students in high school and after graduating when continuing their studies in collage. Based on the results of interviews the problems that often occur there are some students who want to change majors in the middle of the semester, this is because of students cannot follow their lessons and feel left behind by their friends. Therefore, we need an intelligent system that can facilitate the school in grouping students into majors in accordance with the interests and talents of students. In this research the system was made by applying the K-NN method and genetic algorithm optimization. The type of validation used in this research utilizes 9-fold cross validation and hold-out validation. The number of datasets which originally consisted of 288 data will be divided into 9 sections and each sections will amount to 32 data. In general, the best fold number to use is 10, but the share of fold must also be adjusted to the amount of data used. The hold-out test is divided into 2 test scenarios, the first is testing uses the polynomial kernel formula, the RBF kernel and the linear kernel which are elaborated (substituted into the the elaborated distance formula) get a fitness value of 64.338% while the second is testing uses the polynomial kernel formula, kernel RBF and linear kernel which are not elaborated (without substituted into the elaborated distance formula) get a fitness value of 93.182%. The highest fitness value is generated in the 9-fold cross validation test which is 100%.
Co-Authors Imam Cholissodin Putra Pandu Adikara