GraphNews

Knowledge management is already happening across your organization. It’s just happening in silos. This was one of the most important insights I took away from KM World last year in D.C. Many functions are building taxonomies. Many teams are defining hierarchies. Ontologies are emerging in pockets and various applications across the enterprise. But the metadata underpinning them - customers, products, locations, suppliers - often lacks consistent, shared definitions. This wouldnt be a problem if the data only lived in these individual silos of KM, but it doesnt. It's data being shared widely, for both analytical and operational purposes. So what do we really have? Fragmented meaning. I believe knowledge management is the key to unlocking the latent value trapped in unstructured data - and to governing that data safely as we expose it to AI agents. And if 2026 truly becomes “the year of the ontology,” as many predict, this issue will only become more urgent. But here’s the problem. KM today is largely happening inside functional silos, with limited cross-functional governance and little coordination with enterprise data management. At the same time, we lack shared governance policies and common definitions for the metadata that spans both structured and unstructured environments - including for our shared master data. Ironically, we can’t even agree on the terminology for how we capture definitions - controlled vocabularies? data definitions? business glossaries? We are trying to scale AI without first scaling shared meaning. So yes - context matters. But a lack of any cohesive management over the data and processes which help us to understand meaning - and more importantly - integrate that meaning into the systems and processes supporting AI, is a massive problem. So how do we fix it? How do data and knowledge leaders converge around shared governance, shared semantics, and shared accountability? I’d love to hear how others are approaching this. #knoweldgemanagement #KM #datamanagement | 25 comments on LinkedIn

𝗧𝗵𝗲𝗿𝗲'𝘀 𝗻𝗼 𝘄𝗮𝘆 𝘁𝗼 𝘃𝗮𝗹𝗶𝗱𝗮𝘁𝗲 𝗮 𝗚𝗤𝗟 𝗾𝘂𝗲𝗿𝘆 𝘄𝗶𝘁𝗵𝗼𝘂𝘁 𝘀𝗲𝗻𝗱𝗶𝗻𝗴 𝗶𝘁 𝘁𝗼 𝗮 𝗱𝗮𝘁𝗮𝗯𝗮𝘀𝗲.

Vibe coding tools are powerful, but most are not aware of your entire codebase structure.

The other day I found Alan Pike’s blog post from a few years ago which describes his iterative process for determining the order cities should be closed in the game Two Spies. Ultimately, finding a formal solution to city closing wasn’t necessary for the game, but it’s worth giving it a shot anyways since pruning by hand isn’t always convenient.

Do LeBron James and I have a lot of friends in common? Instagram might tell you that we do. After all, many of the people who follow me on Instagram also follow LeBron. So logically, someone who follows Lebron might… Read more →

When and How to Implement Text2Cypher in Agentic Applications Courtesy of Microsoft Designer Introduction In this article, we’ll discuss the process of generating database queries from natural language with LLMs, as well as some of the challenges that come along… Read more →

If you can navigate the complex social dynamics of a high school cafeteria, then you can understand how Louvain works. Imagine it’s your first day at a new high school. You’re walking through the cafeteria, trying to get the lay… Read more →

How relational-era ORM patterns quietly sabotage graph-powered apps — and what to use instead

SurrealDB Raises $23M Series A extension to power the AI-native database era

Learn what exactly is GraphRAG, the three common question types GraphRAG systems faces, and why Atomic GraphRAG’s single query execution layer is truly a game changer.

A practical walkthrough using a real-world Diabetes Patient Risk Stratification system built with LangGraph, Neo4j, PostgreSQL, OWL…

I built a tool that turns your AI coding sessions into a searchable knowledge graph.

When not creating a taxonomy from scratch, the options of re-using a taxonomy, licensing a taxonomy, or using AI to generate a taxonomy are compared.

The Untapped Value of Dark Data

This year's cohort of clients at hypergraf.ai face very similar problems although their domains and approaches are dramatically different.

AI writes code, but lacks production context. Context graphs capture decision traces to build real world models.

In the rapidly evolving landscape of artificial intelligence, the role of symbolic AI—particularly ontologies, RDF semantic knowledge graphs, and inference engines—is often overlooked. However, these components form a critical foundation for building trustworthy AI systems, especially when combined

🚀 Exciting News in Data & Analytics 🚀

🔎 Finding Fraud Rings in a Sea of Transactions: A Graph Data Science Approach

Gartner just made it the #1 infrastructure priority. Now the question is how you get there.

W3C Context Graph Community Group

One of the best works I’m most proud of in my PhD study: BaLu (named in memory of my beloved cat who passed away two years ago).

I have realised there is a need for a part 0 of From Data Engineering to Knowledge Engineering.

A single-pass pattern for AI-assisted semantic web development

𝗪𝗵𝘆 𝗗𝗴𝗿𝗮𝗽𝗵 ‘𝗙𝗮𝗶𝗹𝗲𝗱’ 𝗮𝘀 𝗮 𝗕𝘂𝘀𝗶𝗻𝗲𝘀𝘀 (𝗕𝘂𝘁 𝗜𝘁𝘀 𝗩𝗮𝗹𝘂𝗲 𝗝𝘂𝘀𝘁 𝗚𝗼𝘁 𝗕𝗶𝗴𝗴𝗲𝗿)

🕸️ What if hallucination detection was a graph problem?

A Practical Guide to Transforming Relational Data into Semantic Knowledge Graphs

Ontology Design Principles 10/10: Usable vs Complete

Operationalising Graph Semantics in AI-Driven Systems