Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal RABIT: Jurnal Teknologi dan Sistem Informasi Univrab
Nida Nur Aini Aryanti
Universitas Mercu Buana Yogyakarta
Computer Science & IT Engineering

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

Found 1 Documents
Search

 1 
ANALISIS SENTIMEN TERHADAP PEMUTUSAN HUBUNGAN KERJA DI INDONESIA : KOMPARASI INDOBERT DENGAN SVM, RANDOM FOREST, DAN DECISION TREE DENGAN OPTIMASI TF - IDF Nida Nur Aini Aryanti; Ozzi Suria
Rabit : Jurnal Teknologi dan Sistem Informasi Univrab Vol 10 No 2 (2025): Juli
Publisher : LPPM Universitas Abdurrab

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36341/rabit.v10i2.6364

Abstract

Employment termination (PHK) has become a crucial issue widely discussed on social media platforms like Twitter due to its social and economic impacts. This study aims to analyze public sentiment toward PHK in Indonesia using four classification methods: IndoBERT, Support Vector Machine (SVM), Random Forest, and Decision Tree. The data were collected via Twitter crawling using the keyword "PHK” in Bahasa Indonesia during the period from January to May 2025, resulting in 36,507 tweets. The data underwent preprocessing steps including case folding, cleaning, tokenization, normalization, stopword removal, and stemming. Text features were transformed into numerical form using the TF-IDF for classical models and IndoBERT tokenizer for the transformer-based model. Sentiments were classified into positive, negative, and neutral categories. Model performance was evaluated using a Confusion Matrix with an 80% training and 20% testing data split. Results show that the IndoBERT achieved the highest accuracy at 89,6%, SVM algorithm achieved accuary 88%, precision 90%, recall 95%, F1-score 92%, followed by Random Forest at 78.04%, and Decision Tree at 70.40%. Negative sentiment dominated with 21,790 tweets, reflecting significant public concern over PHK policies. This study concludes that SVM with the TF-IDF approach is the most effective model for real-time public sentiment classification. The limitation of this research lies in the data source, which is exclusively from Twitter and limited to a specific time frame, thus not representing the overall public opinin comprehensively.
Co-Authors Ozzi Suria