EUVIMEDEUVIMED entdecken
Crossref

Self-verification-based framework for key information extraction from unstructured text: A cybercrime investigation support perspective

ByungHui Yoon, Jion Kim

Journal of Data Forensics Research · 31.12.2025 · 2 (2) · 143-159

DOI: 10.12972/jdfr.2025.2.2.2 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Data Forensics Research, online erschienen am 31.12.2025. Verfasst von ByungHui Yoon und 1 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

With the continuous increase in the type and number of online cybercrimes, the volume of unstructured text data that must be analyzed for investigations is increasing. To reduce the workload of investigators handling this massive amount of data, automated methods are urgently required to rapidly extract the key information required for criminal investigations. However, building domain-specialized large language models (LLMs) through fine-tuning entails substantial costs, and cloud-based models face practical limitations in real-world investigative settings owing to the risk of leaking sensitive information. Additionally, when LLMs are used as-is without additional safeguards, errors such as hallucinations hinder ensuring the reliability of their outputs. To address these challenges, this study employs security-preserving on-premise LLMs to extract ten categories of key information from Korean unstructured text messages and proposes a self-verification-based validation procedure to improve the extraction accuracy. We compared the F1 scores before and after applying the verification stage across the four types of on-premise models and observed performance improvements for most of them. These findings suggest that, even under stringent security constraints in investigative environments, combining on-premise LLMs with a systematic verification pipeline can achieve higher accuracy in extracting the core information necessary to support cybercrime investigations.

Zitieren

APA

Yoon, B., & Kim, J. (2025). Self-verification-based framework for key information extraction from unstructured text: A cybercrime investigation support perspective. Journal of Data Forensics Research, 2(2), 143-159. https://doi.org/10.12972/jdfr.2025.2.2.2

Vancouver

Yoon B, Kim J. Self-verification-based framework for key information extraction from unstructured text: A cybercrime investigation support perspective. Journal of Data Forensics Research. 2025;2(2):143-159. doi:10.12972/jdfr.2025.2.2.2

BibTeX
@article{Yoon2025,
  author = {Yoon, ByungHui and Kim, Jion},
  title = {Self-verification-based framework for key information extraction from unstructured text: A cybercrime investigation support perspective},
  journal = {Journal of Data Forensics Research},
  year = {2025},
  volume = {2},
  number = {2},
  pages = {143--159},
  doi = {10.12972/jdfr.2025.2.2.2}
}
Kostenlos starten →