EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Analisis Sentimen Masyarakat Terhadap Kebocoran Pusat Data Nasional Sementara Menggunakan Algoritma Random Forest dan Support Vector Machine

Faishal Khairi Basri, M Afdal, Angraini Angraini, Nesdi Evrilyan Rozanda

Building of Informatics, Technology and Science (BITS) · 02.09.2025 · 7 (2) · 960-971

DOI: 10.47065/bits.v7i2.7473 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Building of Informatics, Technology and Science (BITS), online erschienen am 02.09.2025. Verfasst von Faishal Khairi Basri und 3 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

A ransomware attack on Indonesia’s Temporary National Data Center (PDNS) in June 2024 triggered major public concern over data security and government preparedness. This study aims to analyze public sentiment toward the incident using an Aspect-Based Sentiment Analysis approach on 2,700 Indonesian-language tweets collected from the X platform. The research follows the SEMMA (Sample, Explore, Modify, Model, Assess) methodology, involving text preprocessing, aspect extraction using part-of-speech tagging and named entity recognition, feature representation using Term Frequency-Inverse Document Frequency, and aspect refinement through semantic coherence. Extracted aspects are grouped into five categories: data security, institutions, infrastructure, politics and economy, and impact. Sentiment classification is carried out using the IndoBERTweet model. Results indicate a strong dominance of negative sentiment, particularly in the infrastructure and institutional categories, with no positive sentiment recorded in the political and economic aspect. To address class imbalance in sentiment distribution, the Synthetic Minority Oversampling Technique is applied during model training. Performance evaluation of two algorithms—Random Forest and Support Vector Machine—shows that Random Forest performs best, achieving 96% accuracy on a 70:30 data split and 99.05% average accuracy using 10-fold cross-validation. These findings highlight the effectiveness of aspect-based sentiment analysis and demonstrate Random Forest's superiority in handling imbalanced sentiment classification tasks.

Zitieren

APA

Basri, F. K., Afdal, M., Angraini, A., & Rozanda, N. E. (2025). Analisis Sentimen Masyarakat Terhadap Kebocoran Pusat Data Nasional Sementara Menggunakan Algoritma Random Forest dan Support Vector Machine. Building of Informatics, Technology and Science (BITS), 7(2), 960-971. https://doi.org/10.47065/bits.v7i2.7473

Vancouver

Basri FK, Afdal M, Angraini A, Rozanda NE. Analisis Sentimen Masyarakat Terhadap Kebocoran Pusat Data Nasional Sementara Menggunakan Algoritma Random Forest dan Support Vector Machine. Building of Informatics, Technology and Science (BITS). 2025;7(2):960-971. doi:10.47065/bits.v7i2.7473

BibTeX
@article{Basri2025,
  author = {Basri, Faishal Khairi and Afdal, M and Angraini, Angraini and Rozanda, Nesdi Evrilyan},
  title = {Analisis Sentimen Masyarakat Terhadap Kebocoran Pusat Data Nasional Sementara Menggunakan Algoritma Random Forest dan Support Vector Machine},
  journal = {Building of Informatics, Technology and Science (BITS)},
  year = {2025},
  volume = {7},
  number = {2},
  pages = {960--971},
  doi = {10.47065/bits.v7i2.7473}
}
Kostenlos starten →