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
Indah Ati
Universitas Islam Lamongan
Computer Science & IT Engineering

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

Found 1 Documents
Search

 1 
ANALISIS SENTIMEN PENGGUNA TWITTER TERHADAP KEBIJAKAN PENCAMPURAN ETANOL PADA BBM MENGGUNAKAN METODE NAIVE BAYES Indah Ati; Siti Mujilahwati
Rabit : Jurnal Teknologi dan Sistem Informasi Univrab Vol 11 No 2 (2026): Juli
Publisher : LPPM Universitas Abdurrab

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.36341/rabit.v11i2.8283

Abstract

The Indonesian government's policy on ethanol blending in fuel has triggered public reactions on Twitter (X), particularly complaints about engine sputtering. This study analyzes public sentiment toward the ethanol fuel policy using the Naïve Bayes algorithm with TF-IDF feature extraction. Data were scraped using Apify with five keywords from October 2025 to May 2026, yielding 3,268 raw tweets reduced to 2,137 clean data points. Preprocessing comprised six stages: cleaning, case folding, slangword normalization, tokenizing, stopword filtering, and stemming using the Sastrawi library. Sentiment labeling used an Indonesian Lexicon method, and data were split 80:20 via Stratified Sampling into 1,709 training and 428 testing samples. Labeling showed a dominance of negative sentiment at 59.1% (1,262 data), followed by positive sentiment at 33.8% (723 data) and neutral sentiment at 7.1% (152 data). The Naïve Bayes model achieved an overall accuracy of 75.5%, with the best performance on the negative class (recall 91.3%, F1-Score 82.4%), moderate performance on the positive class (F1-Score 69.2%), but very poor performance on the neutral class (recall 3.3%, F1-Score 6.2%) due to extreme class imbalance. The study concludes that public opinion is predominantly negative, and that handling imbalanced data remains the primary priority for future model development.
Co-Authors Siti Mujilahwati