Vers une mod'elisation de la confiance dans le renseignement sur les menaces cyber

2025-04-02

AI Summary

Assessing and fusing the credibility of Cyber Threat Intelligence (CTI) remains challenging due to heterogeneous, uncertain, and dynamically evolving sources. Method: This paper proposes a multi-valued logic–based trust modeling framework—the first to integrate multi-valued logic into CTI trust assessment—unifying source reliability, information veracity, and logical consistency as orthogonal trust dimensions. It supports unknown parameter handling and dynamic dimension expansion via quantifiable, composable trust representations and aggregation operators. Contribution/Results: The framework enables formal, uncertainty-aware reasoning for cross-source CTI evaluation and knowledge consolidation, significantly enhancing decision robustness and operational adaptability in intelligence-driven defense. It provides both a rigorous theoretical foundation and a scalable methodology for building trustworthy CTI systems.

Technology Category

• Reasoning under Uncertainty: Other Foundations of Reasoning under Uncertainty
• Data Mining & Knowledge Management: Representing, Reasoning, and Using Provenance, Trust
• Machine Learning: Calibration & Uncertainty Quantification

Application Category

• Web Mining and Content Analysis: Web data provenance, reliability, and authenticity
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Cyber threat intelligence (CTI) is essential for effective system defense. CTI is a collection of information about current or past threats to a computer system. This information is gathered by an agent through observation, or based on a set of sources. Building intelligence only makes sense if you have confidence in it. To achieve this, it is necessary to estimate the confidence in each piece of information gathered, taking into account the different dimensions that can make it up: reliability of the source, competence, plausibility of the information, credibility of the information, for example. The information gathered must then be combined with other information to consolidate an agent's knowledge. Recent advances have been made in the theory underlying the modeling of trust for decision-making based on uncertain information, notably by using multivalued logic. This approach makes it possible to deal with unknown values of trust-building parameters, or to easily integrate dimensions. In this article we present the problem of CTI and CTI information sharing, and the reasons that led us to use a logic-based solution for an initial implementation.

Problem

• Modeling confidence in cyber threat intelligence (CTI) for reliable defense
• Combining uncertain CTI data with trust dimensions like source reliability
• Using multivalued logic to handle unknown trust parameters in CTI

Innovation

• Multivalued logic for uncertain information modeling
• Estimating confidence in cyber threat intelligence
• Logic-based solution for CTI information sharing