Network threat intelligence multi-dimensional IoC entity identification method and device based on association enhancement

The invention discloses a network threat intelligence multi-dimensional IoC entity identification method and device based on association enhancement, and the method comprises the steps: forming a multi-dimensional IoC entity unified description language and an abstract standard through employing a s...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: WANG JUNFENG, YU ZHONGKUN, YU JIAN, TANG BINHUI, GE WENHAN, CHEN BAIHAN
Format: Patent
Sprache:chi ; eng
Schlagworte:
Online-Zugang:Volltext bestellen
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:The invention discloses a network threat intelligence multi-dimensional IoC entity identification method and device based on association enhancement, and the method comprises the steps: forming a multi-dimensional IoC entity unified description language and an abstract standard through employing a semi-supervised entity decomposition and relation unified description method for the multi-dimensional complexity of network threat intelligence text data; and adopting a deep learning algorithm combining association enhancement and transfer learning to carry out identification, extraction, verification, evaluation and optimization on the multi-dimensional IoC entities and entity relationships. According to the method, multi-dimensional semantic characteristics are greatly enriched through an implicit relation superposition transfer module and an attention mechanism, meaningless information interference is reduced, additional attributes such as implicit states lost in a traditional method are beneficially combined i