Article Per Year (5 Year)
p-Index From 2019 - 2024
0.23
P-Index
This Author published in this journals
All Journal Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Firhad Rinaldi Saputra
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 Isu Suku, Antar Golongan, Ras, Agama (SARA) pada Twitter Berbahasa Indonesia menggunakan Metode Improved K-Nearest Neighbor (K-NN) Firhad Rinaldi Saputra; Indriati Indriati; Sutrisno Sutrisno
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 4 No 1 (2020): Januari 2020
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (489.941 KB)

Abstract

Twitter is a social network that has one of the most active users today. With the openness of information users move to send texts or tweets about other users, the number of Twitter users makes a lot of tweets related to ethnic issues, between groups, races, religions (SARA). Twitter cannot access the content of tweets that contain Sara's Issues, research is needed to classify tweets to understand including categories of Sara's Issues or Not Sara's Problems. Classification The Sara issue starts in several ways, namely preprocessing which consists of several stages, namely cleaning, folding cases, tokenisation, filtering and stemming. Followed by the term weghting process, to the classification process using the Improved K-Nearest Neighbor method. Based on the implementation and testing carried out in the research on Sara's Issue Classification on Twitter Using K-NN Increase, get the best results based on Precision averages of 0.976422, Remember at 1, F-Measure of 0.987944444 and Accuracy of 96%. Where the number of documents used as training data are 320 documents and test data as many as 80 documents. Where the number of documents, comparison or balance of training data and the value of k-value used determine the good or not classification process of the document.
Co-Authors Indriati Indriati Sutrisno Sutrisno