Image of PENDEKATAN NAMED ENTITY RECOGNITION TERHADAP IOB1 DAN IOB2 TAGGING SCHEME UNTUK EKSTRAKSI ENTITAS PADA CYBER THREAT INTELLIGENCE MENGGUNAKAN MODEL BILSTM-CRF
Bookmark Share

Skripsi

PENDEKATAN NAMED ENTITY RECOGNITION TERHADAP IOB1 DAN IOB2 TAGGING SCHEME UNTUK EKSTRAKSI ENTITAS PADA CYBER THREAT INTELLIGENCE MENGGUNAKAN MODEL BILSTM-CRF

Cyber Threat Intelligence (CTI) is essential to support cyber threat detection and mitigation, particularly for Advanced Persistent Threat (APT) activities that are commonly reported in unstructured text. This condition makes critical information difficult to utilize automatically without an entity extraction process. This study aims to analyze the performance of Named Entity Recognition (NER) for entity extraction from APT reports in CTI and to examine the impact of using the IOB1 and IOB2 tagging schemes on model performance. The proposed method employs a BiLSTM-CRF model with two tagging scheme scenarios, namely IOB1 and IOB2. The dataset used is CyberNER, consisting of 6,311 sentences and 204,815 words. The evaluation was conducted using accuracy, precision, recall, and F1-score metrics. The results indicate that the model achieves high and consistent performance. Using the IOB1 tagging scheme, the model obtained an accuracy of 96.00%, precision of 95.35%, recall of 96.00%, and an F1-score of 95.56%. Using the IOB2 tagging scheme, the model obtained an accuracy of 96.02%, precision of 95.35%, recall of 96.02%, and an F1-score of 95.55%. Although the performance difference between the two tagging schemes is not significant, IOB2 shows more consistent entity boundary labeling across several entity classes. Overall, these findings demonstrate that the BiLSTM-CRF-based NER approach is effective for entity extraction in the CTI domain.

Availability
#
Central Library (Reference) T1933302026
T193330
Available but not for loan - Not for Loan
Detail Information
Series Title
-
Call Number
T1933302026
Publisher
Indralaya : Prodi Sistem Komputer, Fakultas Ilmu Komputer Universitas Sriwijaya.,
Collation
xiv, 48 hlm.; ilus.; 29 cm
Language
Indonesia
ISBN/ISSN
-
Classification
006.350 7
Content Type
Text
Media Type
unmediated
Carrier Type
-
Edition
-
Subject(s)
Prodi Sistem Komputer
Keamanan Siber – Pengolahan Data Teks
Specific Detail Info
-
Statement of Responsibility
Other version/related
File Attachment
Comments
You must be logged in to post a comment