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
Tanica Rakasiwi
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 Artikel Publikasi berdasarkan Judul pada Jurnal Teknologi Informasi dan Ilmu Komputer (JTIIK) Universitas Brawijaya dengan menggunakan Metode Improved K-Nearest Neighbor Tanica Rakasiwi; Bayu Rahayudi; Achmad Ridok
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 5 No 10 (2021): Oktober 2021
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

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

Abstract

Publication articles are contained in journals published by universities or colleges and many are also published by institutions providing national and international journals. Journal of Information Technology and Computer Science (JTIIK) is one of several journals that accept publication articles based on the principle of open dissemination of publications to advance science, especially in the IT field. JTIIK itself is managed by the Faculty of Computer Science, Universitas Brawijaya and started publishing journals in 2014. At JTIIK, the author categorises Publication Articles manually by the author when submitting the manuscript or by the editor during the review and editing process. Therefore, there is a need for automatic grouping of Published Articles documents. The focus of this study is to determine the application of the Improved K-Nearest Neighbor method in the classification of the title of JTIIK Publication Articles. This research was carried out in several processes: Preprocessing, Word Weighting, Feature Selection, Cosine Similarity, and classification with Improved K-Nearest Neighbor. Testing the value of k and using feature selection with K-Fold Cross Validation, the results of changing the value of k and the application of feature selection affect the evaluation results of Improved K-Nearest Neighbor, with a value of k = 5 and feature selection with a threshold DF = 1 as the optimal parameter. Tests with all data resulted in Accuracy of 90.00%, F-measure of 85.78%, Recall of 87.58% and Precision of 84.40%.
Co-Authors Achmad Ridok Bayu Rahayudi