Handling Uncertainty and Vagueness in Network Knowledge Representation for Cyberthreat Intelligence

The overwhelming and constantly growing amount of data network analysts have to process urges automated mechanisms for cyberthreat intelligence. The formal representation of network knowledge unifies data from diverse sources, and enables efficient automation via reasoning, however, network data is inherently uncertain and/or imprecise. There are various approaches to capture data certainty and vagueness, both at the level of abstraction and implementation, but many of these are not decidable, diverge from standards, and are limited in terms of querying and inference support. This paper proposes a description logic formalism to represent uncertain and fuzzy cyberknowledge while keeping favorable computational properties and implementation concerns in mind.