Article Per Year (5 Year)
p-Index From 2021 - 2026
0.659
P-Index
This Author published in this journals
All Journal MATICS : Jurnal Ilmu Komputer dan Teknologi Informasi (Journal of Computer Science and Information Technology) Jurnal Abdi Insani Journal of Applied Data Sciences
Savitri, Ratna
Unknown Affiliation
Agriculture, Biological Sciences & Forestry Civil Engineering, Building, Construction & Architecture Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Education Engineering Environmental Science Health Professions Languange, Linguistic, Communication & Media Law, Crime, Criminology & Criminal Justice Mathematics

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Implementation of BERT in Sentiment Analysis of National Digital Samsat (SIGNAL) User Reviews Based on Machine Learning Savitri, Ratna; Rizki, Fido; Sobri, Ahmad
MATICS: Jurnal Ilmu Komputer dan Teknologi Informasi (Journal of Computer Science and Information Technology) Vol 17, No 2 (2025): MATICS
Publisher : Department of Informatics Engineering

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.18860/mat.v17i2.32059

Abstract

The SIGNAL application facilitates online vehicle tax payments for the public. The application's quality is frequently evaluated through user reviews on platforms like the Google Play Store. This study aims to analyze the sentiment of SIGNAL user reviews using a Machine Learning-based approach, specifically the BERT (Bidirectional Encoder Representations from Transformers) model. The dataset consists of 20,000 user reviews. After preprocessing, the remaining data comprises 17,287 reviews, categorized into 12,758 positive reviews, 2,160 neutral reviews, and 2,369 negative reviews. To address data imbalance, the Random Over Sampling (ROS) technique was applied. The evaluation was performed using metrics such as accuracy, precision, recall, and F1-score. The results of the study indicate that the IndoBERT model can classify sentiments with an accuracy of 99% and a validation accuracy of 98% after five epochs of training. Confusion matrix analysis shows that the model achieved an overall accuracy of 99.72% on training data and 98.68% on testing data. This study demonstrates that the IndoBERT model is highly effective in classifying sentiment and makes a significant contribution to understanding the user experience of SIGNAL, which can serve as a foundation for future improvements to the application.
Co-Authors Apriyana, Faidah Ashari, Devisa Febrilia Bestziel, Jheardy Bulan, Dewi Embong Dworo, Dworo Fazar, Reza Nur Firmansyah Firmansyah Ibrahim Ibrahim Khumairoh, Ai Syarifah M Said Hasibuan Margareta, Fira Santi Oktavianingsih, Okky Rizki, Fido S, Riski Alfira Wulandari Septiana, Yessi sobri, ahmad Toufik, Julianti Fitria