For all the presenters, note that the timer (your allocated time- includes the Q&A time, so make sure that you pay attention to the timer on the screen
", "time": "2025-11-05T19:42:21.000Z"}, {"author": "Robert Wilton", "text": "Chongfeng, when you said 930 AI agents. Are all of these AI agents doing different roles/configuration, or are some of these agents the same, and you have a higher number for scale/performance reasons.
", "time": "2025-11-05T19:52:05.000Z"}, {"author": "Robert Wilton", "text": "Oh, and thanks for a great presentation!
", "time": "2025-11-05T19:52:52.000Z"}, {"author": "Thomas Graf", "text": "@Chongfeng, Can you detail what does data quality mean in this context. Are you referring to schema and semantics? Or delay and loss of data? Or data integrity?
", "time": "2025-11-05T19:53:10.000Z"}, {"author": "Kent Watsen", "text": "@thomas you\u2019re still in queue?
", "time": "2025-11-05T19:53:50.000Z"}, {"author": "Thomas Graf", "text": "@Chongfeng and @Dan, To detail on the points on knowledge graph, Agentic AI and LLM's. The goal of Network Anomaly Detection is fault identification, impact scope analysis and identifying the causality chain. In https://datatracker.ietf.org/doc/html/draft-ietf-nmop-network-anomaly-architecture-05#section-2.3 we are describing the principle of knowledge based detection. In https://www.linkedin.com/pulse/agentic-ai-its-network-analytics-applicability-thomas-graf-kirle/ I am describing the importance of structured data and its semantics, how ontology help to interact between humans and machine, knowledge graphs to store knowledge and establish relationships and how LLM's with Agentic AI can make use of it. In https://datatracker.ietf.org/doc/html/draft-ietf-nmop-network-anomaly-lifecycle we describe how that knowledge can be refined. Therefore I see that the next step in Network Anomaly Detection is to detail the scope of the Postmortem system. Besides operational and analytical metrics which knowledge needs to be stored, linked among and graphed as well. From there we can explore how that information can be used with Agentic AI to refine and optimize the detection rules and the symptom and causality information.
", "time": "2025-11-05T19:56:18.000Z"}, {"author": "Italo Busi", "text": "I do not think having a programmatic solution it is a blocking issue but a facilitator
", "time": "2025-11-05T20:18:30.000Z"}, {"author": "Dan Voyer", "text": "thanks @Thomas, good reminder - @Chongfeng I'm still curious to see how you lined up your knowledge graph, I'm not challenging - just trying to see if you had another approach
", "time": "2025-11-05T20:18:41.000Z"}, {"author": "Italo Busi", "text": "Implementations already exist with no need for a programmatic solution
", "time": "2025-11-05T20:19:25.000Z"}, {"author": "Mahesh Jethanandani", "text": "Maybe this is a stupid question, but I am trying to understand what the issue is. Let us say the model has 10 leafs, but a particular use case only needs 5. Although not recommended, can you not 'deviate not-implemented'?
", "time": "2025-11-05T20:24:08.000Z"}, {"author": "Italo Busi", "text": "RFC9656 has already been published providing technology specific augmentations for a very simple profile of RFC8795
", "time": "2025-11-05T20:24:20.000Z"}, {"author": "Italo Busi", "text": "See section 2.4 of RFC9656
", "time": "2025-11-05T20:24:42.000Z"}, {"author": "Reshad Rahman", "text": "@Mahesh and all, the profile document is https://datatracker.ietf.org/doc/draft-ietf-teas-te-topology-profiles/
", "time": "2025-11-05T20:26:57.000Z"}, {"author": "Boris Khasanov", "text": "Bye!
", "time": "2025-11-05T20:32:09.000Z"}]