Building an IT Service Management Knowledge Graph
summary
Built a self-refreshing ITSM knowledge graph inside Freshservice that auto-classifies tickets at scale by combining clustering to surface unknown terms, sampling plus manual annotation, and retraining a sentence-encoder NER + span classifier. System processes millions of tickets, auto-maps known items (Cisco VPN → VPN security product → higher priority) and flags long‑tail misses like “Photoshop.” GPT-assisted sampling speeds discovery, but ontology edits and full rollouts remain semi‑manual and are applied quarterly.
AI-written summary · about
chapters (6)
- 0:00Introduction to knowledge graphs
- 3:12Why knowledge graphs matter
- 8:30Building the graph structure
- 13:50Discovering new knowledge
- 23:04Manual updates and maintenance
- 24:08Q&A and closing
related talks
24mYour company, as a Knowledge Graph - the foundation of cybersecurity’s future
33mSmarter Industrial Fault Diagnosis with Knowledge Graphs and Large Language Models
44mBuilding Knowledge Graphs- 44mBuilding Knowledge Graphs
15mGraphRAG-Powered AI Agents: Transforming Corporate Metadata into Intelligent Knowledge
16mMedQGraph: TMKGs for AI-Driven Healthcare Insights
28mData-Empowered AI Assistants
42mBuilding a knowledge graph for climate policy