This invention describes a system for monitoring computer networks and services. It automatically groups related events that happen in the system, figuring out these groupings on its own without needing someone to write specific rules. A user can interact with a display to tell the system to focus on monitoring events for particular services, allowing it to predict and prevent problems by sending corrective messages. The claims specifically focus on a user interface for controlling this by-service mode and then automatically classifying events into predefined "seed groups" based on identified factors.
Why it matters: The core of this invention relies on automatically determining event groupings and classifying events based on identified factors without explicit rules. Since its filing in 2019, advancements in machine learning, particularly in areas like unsupervised learning, anomaly detection, and large language models, have made such dynamic and intelligent event correlation significantly more feasible and robust by 2026.
AI gives you a few directions you could take this. Pick one, and we check whether your version is different enough to patent, then write the filing.
Reinvent this with AI