Perbandingan Semantic Search Berbasis Sentence-BERT dan TF-IDF pada Pencarian Dokumen Ancaman Siber
Keywords:
semantic search, Sentence-BERT, TF-IDF, pencarian dokumen, ancaman sibeAbstract
Peningkatan jumlah dokumen ancaman siber menyebabkan proses pencarian informasi menjadi semakin kompleks karena metode berbasis kata kunci belum mampu memahami hubungan semantik antar dokumen. Penelitian ini bertujuan membandingkan performa metode Term Frequency–Inverse Document Frequency (TF-IDF) dan Sentence Bidirectional Encoder Representations from Transformers (Sentence-BERT) pada sistem pencarian dokumen ancaman siber. Tahapan penelitian meliputi impor dataset CyberBERT, preprocessing teks, pembangunan model pencarian, pengujian menggunakan kueri, serta evaluasi hasil berdasarkan relevansi dokumen dan waktu respons. Dataset yang digunakan terdiri atas 33 dokumen yang diproses menjadi representasi vektor menggunakan TF-IDF dan embedding berdimensi 384 menggunakan Sentence-BERT. Hasil penelitian menunjukkan bahwa kedua metode mampu mengembalikan dokumen yang relevan terhadap kueri yang diberikan, namun menghasilkan urutan dokumen yang berbeda karena perbedaan pendekatan representasi teks. Metode TF-IDF memiliki waktu respons lebih cepat, yaitu 0,0085 detik, sedangkan Sentence-BERT membutuhkan 0,0357 detik. Di sisi lain, Sentence-BERT lebih mampu menangkap hubungan semantik sehingga menghasilkan pencarian yang lebih kontekstual. Dengan demikian, TF-IDF lebih unggul dari sisi efisiensi, sedangkan Sentence-BERT lebih efektif dalam memahami makna dokumen sehingga sesuai untuk sistem semantic search pada domain ancaman siber.
Downloads
References
Z. Hu, H. Dai, G. Yang, X. Yi, and W. Sheng, ‘Semantic-Based Multi-Keyword Ranked Search Schemes over Encrypted Cloud Data’, Security and Communication Networks, vol. 2022, pp. 1–15, Apr. 2022, doi: 10.1155/2022/4478618.
R. M. Holis, P. E. P. Utomo, and B. F. Hutabarat, ‘Semantic FAQ Chatbot Using SBERT (Sentence-BERT) and Cosine Similarity for Academic Services’, Brilliance, vol. 5, no. 2, pp. 915–922, Oct. 2025, doi: 10.47709/brilliance.v5i2.7027.
S.-S. Chen, T.-W. Pai, and C.-Y. Sun, ‘EnhanceCTI: Enhanced semantic filtering and feature extraction framework for industry-specific cyber threat intelligence’, Computers & Security, vol. 158, p. 104649, Nov. 2025, doi: 10.1016/j.cose.2025.104649.
Z. Hu, H. Dai, G. Yang, X. Yi, and W. Sheng, ‘Semantic-Based Multi-Keyword Ranked Search Schemes over Encrypted Cloud Data’, Security and Communication Networks, vol. 2022, pp. 1–15, Apr. 2022, doi: 10.1155/2022/4478618.
R. M. Holis, P. E. P. Utomo, and B. F. Hutabarat, ‘Semantic FAQ Chatbot Using SBERT (Sentence-BERT) and Cosine Similarity for Academic Services’, Brilliance, vol. 5, no. 2, pp. 915–922, Oct. 2025, doi: 10.47709/brilliance.v5i2.7027.
S.-S. Chen, T.-W. Pai, and C.-Y. Sun, ‘EnhanceCTI: Enhanced semantic filtering and feature extraction framework for industry-specific cyber threat intelligence’, Computers & Security, vol. 158, p. 104649, Nov. 2025, doi: 10.1016/j.cose.2025.104649.
I. Lubis, H. Lubis, and I. Nur Wahidah, ‘Implementation of Semantic Search in an Academic Repository Using Sentence-BERT and FAISS’, SinkrOn, vol. 10, no. 2, pp. 1060–1069, Apr. 2026, doi: 10.33395/sinkron.v10i2.15940.
M. A. H. Fathuddin, E. P. Mandyartha, and A. L. Nurlaili, ‘Penerapan Sentence-Bert dan Cosine Similarity untuk Pencarian Semantik Dokumen Skripsi dalam Format PDF’, RRJ, vol. 8, no. 1, pp. 322–337, Oct. 2025, doi: 10.38035/rrj.v8i1.1865.
H. Gusdevi, A. Setyanto, K. Kusrini, and E. Utami, ‘Cosine Similarity-Based Evidences Selection for Fact Verification Using SBERT on the FEVER Dataset’, CogITo Smart Journal, vol. 11, no. 1, pp. 52–66, Jun. 2025, doi: 10.31154/cogito.v11i1.917.52-66.
J. Baek, J. O, S. Jeong, and W. Kim, ‘A Study on Improving the Automatic Classification Performance of Cybersecurity MITRE ATT&CK Tactics Using NLP-Based ModernBERT and BERTopic Models’, Electronics, vol. 14, no. 22, p. 4434, Nov. 2025, doi: 10.3390/electronics14224434.
R. Othman, B. Rossi, and R. Barbara, ‘A Comparison of Vulnerability Feature Extraction Methods from Textual Attack Patterns’, 2024, arXiv. doi: 10.48550/ARXIV.2407.06753.
T. Chakraborty, V. La Gatta, V. Moscato, and G. Sperlì, ‘Information retrieval algorithms and neural ranking models to detect previously fact-checked information’, Neurocomputing, vol. 557, p. 126680, Nov. 2023, doi: 10.1016/j.neucom.2023.126680.
A. L. Alfeo, M. G. C. A. Cimino, and G. Vaglini, ‘Technological troubleshooting based on sentence embedding with deep transformers’, J Intell Manuf, vol. 32, no. 6, pp. 1699–1710, Aug. 2021, doi: 10.1007/s10845-021-01797-w.
A. Esteva et al., ‘COVID-19 information retrieval with deep-learning based semantic search, question answering, and abstractive summarization’, npj Digit. Med., vol. 4, no. 1, p. 68, Apr. 2021, doi: 10.1038/s41746-021-00437-0.
C. Min, Y. Chu, L. Yang, B. Xu, and H. Lin, ‘Locality Preserving Sentence Encoding’, in Findings of the Association for Computational Linguistics: EMNLP 2021, Punta Cana, Dominican Republic: Association for Computational Linguistics, 2021, pp. 3050–3060. doi: 10.18653/v1/2021.findings-emnlp.262.
R. Al Rasyid and D. H. U. Ningsih, ‘Penerapan Algoritma TF-IDF dan Cosine Similarity untuk Query Pencarian Pada Dataset Destinasi Wisata’, jtik, vol. 8, no. 1, pp. 170–178, Jan. 2024, doi: 10.35870/jtik.v8i1.1416.
N. Reimers and I. Gurevych, ‘Making Monolingual Sentence Embeddings Multilingual using Knowledge Distillation’, in Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), Online: Association for Computational Linguistics, 2020, pp. 4512–4525. doi: 10.18653/v1/2020.emnlp-main.365.
G. H. Setiawan and I. M. B. Adnyana, ‘Improving Helpdesk Chatbot Performance with Term Frequency-Inverse Document Frequency (TF-IDF) and Cosine Similarity Models’, JAIC, vol. 7, no. 2, pp. 252–257, Dec. 2023, doi: 10.30871/jaic.v7i2.6527.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Nanda Setiawan; Johan, Iswanto, Roslina
This work is licensed under a Creative Commons Attribution 4.0 International License.
Universitas Harapan Medan
.png)
.png)
.png)
.png)
.png)
