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
Rizky Ramadhan
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 
Rekomendasi Pengambilan Judul Skripsi Menggunakan PSO-Neighbor Weighted K-Nearest Neighbor (NWKNN) (Studi Kasus: Jurusan Ilmu Keolahragaan Fakultas Ilmu Keolahragaan Universitas Negeri Medan) Rizky Ramadhan; Imam Cholissodin; Edy Santoso
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 5 No 7 (2021): Juli 2021
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

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

Abstract

This research is non-implementative descriptive with the K-Means technique as the initial formation of the clusters of graduated student thesis titles, the PSO technique as a course selection, and the Neighbor Weighted K-Nearest Neighbor (NWKNN) technique as data classification and algorithm performance measurement using the technique. accuracy. The data collection of this research is in the form of document studies from 2016/2019 graduate students from the Faculty of Sport Sciences, State University of Medan. The purpose of this study was to determine the parameter value and accuracy value of the application of the NWKNN algorithm to provide the best recommendations regarding the thesis title raised by students. The results of this study can be concluded that in testing the percentage of many comparisons of training data and testing data used is 90%: 10%. The generation and feature testing resulted in a generation that began to be constant in the 50th generation and with 15 subjects, namely: MK2, MK6, MK7, MK11, MK12, MK13, MK14, MK15, MK21, MK24, MK26, MK27, MK28, MK31, MK32. In testing the K value, the optimum K value is 3. In testing the K and E values, the optimum K and E values ​​are 3 and 2. The PSO-Neighbor Weighted K-Nearest Neighbor (NWKNN) algorithm for recommendations for thesis title retrieval produces an optimum value using parameters previously obtained an accuracy of 88.28%.
Co-Authors Edy Santoso Imam Cholissodin