PENGEMBANGAN THREAT INTELLIGENCE KNOWLEDGE GRAPH DENGAN ENTITY EXTRACTION TERHADAP ADVANCED PERSISTENT THREAT MENGGUNAKAN PRE-TRAINED DEEPSEEK

ASSYARI, HABIB AL and Stiawan, Deris and Afifah, Nurul (2025) PENGEMBANGAN THREAT INTELLIGENCE KNOWLEDGE GRAPH DENGAN ENTITY EXTRACTION TERHADAP ADVANCED PERSISTENT THREAT MENGGUNAKAN PRE-TRAINED DEEPSEEK. Undergraduate thesis, Sriwijaya University.

Preview	Image RAMA_56201 _09011382126142_cover.jpg - Cover Image Available under License Creative Commons Public Domain Dedication. Download (454kB) | Preview
	Text RAMA_56201 _09011382126142.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (6MB) | Request a copy
	Text RAMA_56201 _09011382126142_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (5MB) | Request a copy
	Text RAMA_56201 _09011382126142_0003047905_0010119206_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (5MB)
	Text RAMA_56201 _09011382126142_0003047905_0010119206_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (740kB) | Request a copy
	Text RAMA_56201 _09011382126142_0003047905_0010119206_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (527kB) | Request a copy
	Text RAMA_56201 _09011382126142_0003047905_0010119206_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (589kB) | Request a copy
	Text RAMA_56201 _09011382126142_0003047905_0010119206_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (178kB) | Request a copy
	Text RAMA_56201 _09011382126142_0003047905_0010119206_06_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (222kB) | Request a copy
	Text RAMA_56201 _09011382126142_0003047905_0010119206_07_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (455kB) | Request a copy

Abstract

This research aims to address complex cybersecurity challenges, namely Advanced Persistent Threat (APT), by developing a Threat Intelligence Knowledge Graph. The study proposes a Natural Language Processing based approach to extract entities and relationships from APT reports, using a pre trained Deepseek model. This model was fine-tuned specifically for Named Entity Recognition (NER) and Relation Extraction tasks. The research findings show that the fine-tuned Deepseek model achieved an F1-score of 0.960, outperforming the BERT model, which only achieved 0.694 under the same conditions and dataset. The primary output of this research is a knowledge graph that effectively visualizes attack entities, such as threat actors, malware, and tactics, into a structured representation that complies with the Structured Threat Information eXpression (STIX) standard. These findings demonstrate that a knowledge graph can be a reliable tool for security analysts to analyze APT attack patterns more quickly and in-depth.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Cyber Threat Intelligence, Threat Intelligence Knowledge Graph, Entity Extraction, Advanced Persistent Threat, Deepseek, Named Entity Recognition, Relation Extraction, Structured Threat Information eXpression.
Subjects:	Q Science > Q Science (General) > Q300-390 Cybernetics > Q325.5 Machine learning Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation.
Divisions:	09-Faculty of Computer Science > 56201-Computer Systems (S1)
Depositing User:	Habib Al Assyari
Date Deposited:	25 Sep 2025 03:05
Last Modified:	25 Sep 2025 03:05
URI:	http://repository.unsri.ac.id/id/eprint/184805

Actions (login required)

View Item