AI-Driven Threat Intelligence: Aggregating Identity Signals for System-Wide Attack Prediction
2025-2026 tavasz
Szoftver
Téma leírása
Current Identity and Access Management (IAM) security systems typically focus on individual user anomalies—flagging if User A is behaving strangely. However, sophisticated attacks often target the system as a whole rather than a single user. Techniques like distributed brute-force attacks, credential stuffing (Account Takeover), or massive account opening fraud only become visible when identity signals are interpolated across the entire user base. There is a market need for a "Threat Intelligence Dashboard" that elevates the view from single-user risk to system-wide "Likelihood of Attack."
The objective of this thesis is to move beyond isolated user analysis and develop a system that aggregates identity signals to detect well-known attack patterns. The student must determine how to correlate separate data points (e.g., failed login rates, IP reputation, geo-location, registration velocity) to calculate a real-time confidence score (a "Gauge") for specific threat scenarios. The system must then suggest actionable mitigations without automatically intervening.
Külső partner: Nevis Security
Maximális létszám:
1 fő