Skripsi

PERBANDINGAN IOB1 DAN IOE TAGGING SCHEME PADA NAMED ENTITY RECOGNITION UNTUK EKSTRAKSI ENTITAS ADVANCED PERSISTENT THREAT MENGGUNAKAN BILSTM

---

Advanced Persistent Threat (APT) reports in Cyber Threat Intelligence (CTI) are generally presented as unstructured text, making them difficult to analyze manually and requiring automated methods to extract important entities quickly and accurately. This study aims to analyze the performance of Named Entity Recognition (NER) using a Bidirectional Long Short-Term Memory (BiLSTM) model with two sequence labeling tagging schemes, namely IOB1 and IOE, for recognizing entities in APT reports. This comparison is important to evaluate the impact of tagging strategies on model performance in identifying entity boundaries within the CTI domain. This study contributes to a comparative evaluation of these two tagging schemes in the CTI domain, which remains relatively underexplored. The dataset used is CyberNER, consisting of 6,311 sentences and 204,815 tokens. The results show that IOB1 achieves an accuracy of 95.65%, precision of 94.62%, recall of 95.65%, and F1-score of 94.87%, while IOE achieves an accuracy of 95.38%, precision of 94.10%, recall of 95.38%, and F1-score of 94.29%. These results indicate that IOB1 outperforms IOE based on the highest F1-score. Overall, the BiLSTM-based NER approach demonstrates good performance, although performance on some minor labels still needs improvement.

---

Availability

Detail Information

Series Title

-

Call Number

T1956712026

Publisher

Indralaya : Prodi Sistem Komputer, Fakultas Ilmu Komputer Universitas Sriwijaya., 2026

Collation

xviii, 56 hlm.; ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

006.350 7

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Prodi Sistem Komputer
Named Entity Recognition

Specific Detail Info

-

Statement of Responsibility

TUTI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment