New CTiKG framework boosts threat intelligence extraction accuracy

Researchers have developed a new framework called CTiKG to improve the extraction of threat entities and their relationships from cybersecurity reports. This framework utilizes a hybrid NLP model that combines SecureBERT contextual embeddings with domain-specific knowledge from an ontology. Experiments show CTiKG achieves significant gains in Named Entity Recognition (NER) and Relation Extraction (RE) performance compared to existing methods, with improvements of 3-4% in NER and up to 8% in RE on a specialized dataset. AI

IMPACT Enhances accuracy in extracting threat intelligence, potentially improving cybersecurity response times and efficiency.

RANK_REASON Academic paper introducing a new framework and demonstrating improved performance on benchmarks. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• CTiKG
• DNRTI
• DNRTI-AUG-STIX2
• SecureBERT
• STIX 2.1

• paper
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →