←
Connected Data6,387 posts
Connected Data avatar

Connected Data

@Connected_Data

Connecting Data, People & Ideas since 2016. Using relationships, meaning, context in Data to achieve great things #KnowledgeGraph #GraphDB #AI #SemTech

◷ Joined April 2016▧ 3,661 Photos and videos
Connected Data avatar
Connected Data@Connected_Data·Nov 17

Connected Data London 2026 takeaways: From AI-ready data to agent-ready enterprise Connected Data London 2026 is a wrap. A massive thank you to the Connected Data London community: the speakers, chairs, panelists and attendees who define the agenda with care, push the field forward and turn ideas into applied innovation. These are some initial takeaways from the 10th edition, 11-12 November. Five attendee write-ups, read side by side, kept landing on the same ground. 🧭 Neural for perception, symbolic for trust LLMs read, translate and structure messy material. Ontologies, rules and shapes cover whatever must be exact, repeatable and explainable. Examples ran from robots learning what a plug entering a socket means, to IKEA's rule that generation never invents a recipe, to a keynote where "don't know" counts as a real output. From AI-ready data to agent-ready enterprise The conversation moved from preparing data for AI to agent memory, context graphs, provenance, evaluation and auditability. Identity, permissions and delegation between agents are questions the community has modelled before: PROV-O has described provenance since 2013. Several voices asked for more working demos that connect semantics to the AI engineering stack. Trust needs identity Governance and human in the loop were reframed as trust and judgement. A decision trace built on a false match is perfectly logged and still leads to a wrong outcome. Knowledge apart from data AstraZeneca separates the model (T-box), reference data (C-box) and data (A-box), mints a unit like eGFR once, and resolves one study carried under five identifier schemes. Juan Sequeda's lessons point the same way, and an ontology derived straight from a database schema can inherit that schema's silos. Standards are hardening SHACL 1.2 is tracked closely. Named graphs are back, with TriG replacing plain Turtle and urn-based partition names. RDF 1.2 reifiers are becoming referenceable. Tooling is maturing too: a package manager for ontologies built on DCAT profiles, and a benchmark of 11 RDF frameworks. People and architecture IKEA's multi-year journey: start small, think big, frame the work as a capability. A younger, AI-native cohort treats ontologies as AI-generated and expert-refined, with competency questions back in focus. Holons appeared in Amazon's session and in life sciences, Markdown with YAML-LD headers is emerging as a messaging format, and most company knowledge still sits in documents. By Kurt Cagle, Jonathan Foxwell, Kerstin Forsberg, Juan Sequeda and Katariina Kari Back home after 36 hours at Connected Data London. A few takeaways based on conversations and talks sbee.link/vnqey6tcku Two days at Connected Data London 2026, co-chairing the Edges track and moderating a panel. These are the ideas I brought home. sbee.link/6bpe9mxft8 Connected Data London 2026 – early report sbee.link/ytf986uv3c Connected Data London 2026: From AI-Ready Data to Agentic Systems sbee.link/n6ved83ukr Home from CDL 2026: The Next Era of the Semantic Stack sbee.link/3kx7wyqvhu Slides and replays coming soon! #CDL2026 #Neurosymbolic #SHACL #ContextGraphs #SemanticWeb -- Missed a talk at #CDL26? Or missed the whole thing? All talk recordings are being processed and will be available in the next few days. With this many ideas and speakers, nobody could catch them all live. We’ve got you covered. ✅ Were you there? Replays are already included in your ticket. 🎟 Couldn’t make it? Get a remote ticket for permanent on-demand access to every talk. 👉 sbee.link/9wbg6arkdp Recorded live at Connected Data London 2026 | 11–12 November | Leonardo Royal Hotel London Tower Bridge #CDL26 #KnowledgeGraph #GraphRAG #Ontology #AI

Post attachment
Connected Data avatar
Connected Data@Connected_Data·2h

Connected Data London post-event goodies: Photo Gallery CDL26 has been a blast, and apparently what's on everyone's minds (and in our inboxes) right now some variation of the following: - Do you have an official gallery? - Can we get access to the speaker slides? - When will the recordings be available? - What's your next event going to be? The answers are yes, yes, any day now, and you'll hear about that one soon! Our AV crew has been roaming the event and is still working on getting all the material processed. In the meanwhile, here's a first taste: the official CDL26 Photo Gallery Look back on your Connected Data London 2026 event experience and get ready for next year! View our tenth anniversary highlights to see how our community collaborated, inspired and rejoiced - our gallery gives you a glimpse of what to expect in 2027. 2026.connected-data.london/g… Stay tuned for slides, videos, and more news!

Post attachmentPost attachmentPost attachmentPost attachment
Connected Data avatar
Connected Data@Connected_Data·6h

Ontologies for AI in practice: Do LLMs need ontologies, or are text definitions good enough? A three-part experiment on a small, deliberately ambiguous genealogy dataset (fictional people) tested what changes when an LLM is given progressively more semantics.  The task: count a person’s ancestors. The evidence is anecdotal by design, but practical. 🔎 Part I: adding semantics step by step Data alone: many errors, and a correct count only by chance Cleaner dates: no change Plain-text definitions: a substantial improvement RDF knowledge graph: the same answer, but it exposed hidden assumptions, including an imprecise question (“how many ancestors in this dataset?”) Curated graph: more precise and consistent answers Ontology: the correct count 🧪 Part II: model size and stability Small local models (4B to 8B) could not use added semantics Impact started to appear at 32B parameters Text definitions produced high variability across runs Knowledge graphs with ontologies matched or beat definitions and were more consistent Removing the text definitions from context improved graph-based results 🧩 Part III: how to provide semantics OWL axioms only, OWL with definitions, OWL with operational directives, Apache Ossie (OSI) and Google’s OKF were compared. Takeaways: - Axioms alone were not enough, and the natural-language definitions inside ontologies did most of the work - With a capable model, the format mattered little. Opus reached perfect results across representations, while DeepSeek-R1:32B used none of them - Ontology plus knowledge graph was the most effective and consistent option - Among approaches that annotate data directly, OKF ranked highest Definitions alone are not enough. Results were strongest when context-specific meaning is connected to the data elements themselves, as in a knowledge graph. By Andrea Splendiani Ontologies for AI in practice / part I sbee.link/gkx3nd6rec Ontologies for AI in practice / part II sbee.link/erhmudfwc6 Ontologies for AI in practice / part III sbee.link/3d7xhwbnm4 #LLM #FAIRData #OWL #SemanticLayer #DataGovernance --      Missed a talk at #CDL26? Or missed the whole thing?      All talk recordings will be available in the next few days.      With this many ideas and speakers, nobody could catch them all live      ✅ Were you there? Replays included in your ticket   🎟 Couldn’t make it? Get a remote ticket for permanent on-demand access sbee.link/pqwxhda3n4

Post attachment
The Year of the Graph avatar
⟳ Connected Data retweeted
The Year of the Graph@TheYotg·Nov 16

LLM wiki and knowledge graph formats and recipes Agent builders are converging on the same discovery: the simplest way to give an agent durable memory is a markdown LLM wiki. Enterprises are converging on semantic layers, context graphs and ontologies to ground their agents. Tony Seale put these together. You can combine a markdown wiki with a formal knowledge graph and ontology, Seale argues. It is an extremely powerful pattern, now published as an open standard for the community to evolve. It’s called Vault-LD. The idea is simple: markdown plus YAML-LD frontmatter. The W3C recently published version 1.0 of the YAML-LD specification. YAML-LD is a set of conventions which specify how to serialize (RDF) Linked Data as YAML based on JSON-LD syntax, semantics, and APIs. JSON-LD is a lightweight, machine-readable format used to store and transport linked data on the web. JSON-LD uses Schema.org vocabularies to explicitly define what web page elements mean (like products, prices, or authors). JSON-LD owns meaning; it’s the format behind the EU Digital Identity Wallet, Microsoft’s Digital Twins Definition Language, and how AI-ready datasets describe themselves on Hugging Face and Kaggle. JSON Schema owns structure; it’s the validation logic behind most modern API specs, MCP tool definitions, and IoT data models. Juan Cruz Viotti addressed JSON Schema and JSON-LD interoperability. Google’s Open Knowledge Format (OKF) formalizes the LLM-wiki pattern. OKF v0.2 adds support for trust signals, to answer questions like “what was this made from?” or “Is it still true?”. Neo4j created ki, an open source project to sync a Neo4j graph deterministically from a folder of markdown, and added unofficial support for OKF. --      📩 Excerpt from The Year of the Graph Autumn 2026 newsletter      Read "# Meaning, Traversal, Reasoning, Action: The Graph Stack for Agentic AI" with more sections, references and attribution here 👇   yearofthegraph.xyz/newslette… All things #KnowledgeGraph, #GraphDB, Graph #Analytics / #DataScience / #AI and #SemTech.

Post attachment
Carlos E. Perez avatar
⟳ Connected Data retweeted
Carlos E. Perez@IntuitMachine·Nov 16

The key perspective I leaned at CDL:

The old ontology was built to make the world intelligible to machines. The new one makes collective intelligence intelligible, and governable, to people. Its purpose is not to hold intelligent systems back. It is to lay the common ground on which different kinds of intelligence can work together. For enterprise architects, that is now the work. Design the loops, the authority, and the commitments. Who holds the loop? Whoever holds the shared meaning, the legitimate authority, and the record of what was agreed. The lasting value of an ontology is not how much intelligence it adds to any one agent. It is how much trustworthy collective agency it makes possible. intuitionmachine.com/explain…

Connected Data avatar
Connected Data@Connected_Data·Nov 17

SPARQL: The RDF Query Language Demystifying Knowledge Graphs, Semantic Technologies and Labelled Property Graphs: A Beginners Hands-On Exploration for making the right choice Knowledge graphs have long underpinned enterprise data strategies, implemented through diverse paradigms ranging from RDF-based semantic technologies to labelled property graphs (LPGs).  While these frameworks have often operated quietly in the background, the rise of large language models (LLMs) has reignited interest in knowledge graphs, bringing long-overdue recognition to their strategic value.  Knowledge graphs not only enhance LLM training and deployment but also offer unmatched capabilities for optimizing data integration, governance, and analytics. Yet, fundamental questions remain: What exactly constitutes a knowledge graph? When should RDF, OWL, or LPGs be applied? Are these technologies interchangeable, complementary, or competitive? This tutorial seeks to demystify these approaches through a pragmatic, hands-on lens. Presenters—bringing distinct perspectives from both RDF and LPG traditions—will guide participants through real-world scenarios, comparative exercises, and applied case studies.  Attendees will gain a nuanced understanding of the trade-offs, synergies, and integration pathways across semantic technologies, property graphs, and their integration with LLMs, equipping them to make informed architectural choices in enterprise and research settings. Participants will explore: - The core principles and unique capabilities of semantic technologies and property graphs. - Practical decision-making frameworks for selecting the right approach for specific use cases. - Insights into emerging trends where these approaches converge to unlock new opportunities. - Considerations to take into account when choosing the tools in order to ensure a smooth implementation.   Learn about some high level strategies for leveraging graph technologies to augment LLMs and drive business value. By the end of the session, attendees will have a deeper understanding of the strengths, limitations and synergies of these graph based approaches, enabling them to make informed decisions that align with their organizational goals when solving different challenges.  Target Audience  Whether you're a technical leader, data scientist, engineer or enterprise strategist, this tutorial will equip you with actionable insights to harness the full potential of graphs in relevance to your use case.  ▶️ 2025.connected-data.london/t… -- Tara Raafat. Head of Metadata and Knowledge Graph Strategy, Bloomberg Dr. Tara Raafat is the head of Metadata and Knowledge Graph Strategy in Bloomberg’s CTO Office, where she leads the development of Bloomberg’s enterprise Knowledge Graph and semantic metadata strategy, aligning it with AI and data integration initiatives. Petra Selmer. Query Language Architect, Bloomberg Dr. Petra Selmer is Bloomberg's Query Language (BQL) Architect and a leading expert in graph query languages. -- Welcome to Connected Data London's #TeaserTuesday Every Tuesday, we share teasers from #CDL25 on our channels Connected Data London 2025 brought together leaders and innovators. Were you there? 🎥 Watch the sessions: sbee.link/rwav6dcngt 📩 Join the community: sbee.link/nh6wb7cdqj Tune in and learn from leaders and innovators; subscribe and watch premieres as they are released!  Join community legends and new voices in #CDL25 for all things #KnowledgeGraph #Graph #analytics #datascience #AI #graphDB #SemTech #Ontology

2:03
The Year of the Graph avatar
⟳ Connected Data retweeted
The Year of the Graph@TheYotg·Nov 13

The Neo4j Ontology Graph Specification: A pragmatic, open, AI-first approach An open-source specification for representing domain knowledge in a machine-readable form, enabling interoperability and automation across data, applications and AI systems. Three design principles: - Pragmatic: entities, relationships, properties and business meaning, described in a form that fits how teams develop, integrate and operate applications. - Interoperable: documented mappings to selected RDFS and SHACL constructs, a reference field linking every element to external standard terms, and added constructs for a wider range of application scenarios. - AI-first: a schema tells an agent what exists, not what it means. Descriptions and aliases tell it that “Account” can be a bank account or a payment account, and that an IBAN identifies it. One ontology, many uses. The same specification can be enforced by the database as constraints, drive ontology-based logic to derive or enrich data, or give applications and agents the context to interpret information. It spans knowledge graph development, GraphRAG, data validation and AI-powered applications. What’s included: - Nodes and relationships with stable IDs, labels, descriptions and aliases - Properties across the full Neo4j type system, including temporal, spatial and vector types - Cardinality and constraints (existence, uniqueness, key, property type) - Named extensions for indexes, display hints and data-source mappings - A single YAML or JSON document, validated against a versioned JSON Schema, with SDKs for JVM, JavaScript and Go - Apache 2.0 licence Similar needs show up in Palantir Foundry, Databricks Genie Ontology and Microsoft Fabric IQ: explicit, machine-readable domain knowledge that data systems and AI can both use. By Jesús Barrasa and Oskar Hane The Neo4j Ontology Graph Specification: A pragmatic, open, AI-first approach neo4j.com/blog/developer/the… Neo4j Ontology Graph Specification neo4j.com/docs/ontology-grap… #Ontology #GraphRAG #Neo4j #AIAgents #OpenSource -- 📩 The Year of the Graph Autumn 2026 newsletter issue is out! Meaning, Traversal, Reasoning, Action: The Graph Stack for Agentic AI. 👇 yearofthegraph.xyz/newslette… All things #KnowledgeGraph, #GraphDB, Graph #Analytics / #DataScience / #AI and #SemTech. Subscribe and follow to be in the know. Reach out if you’d like to be featured Comment link: neo4j.com/blog/developer/the…

Post attachment
Connected Data avatar
Connected Data@Connected_Data·Nov 16

25 years later, the Semantic Web is being reinvented and co-opted..and that's a good thing They say imitation is the sincerest form of flattery. They also say timing is everything. 25 years ago, the Semantic Web gave the world a set of principles and standards that help add meaning and grounding to data. For a long time, the world was more concerned about quantity than quality. Meaning and semantics was overlooked while pipelines and big data were top of mind. Knowledge Graphs and Ontologies, while universally applicable, were reserved for a select few. Today, the world is waking up to the importance of the principles and standards established by the Semantic Web. The technology stack has matured and evolved significantly, with rising adoption driving a new wave of evolution. While the journey has not been easy or flawless, the destination is in sight. James Hendler is one of the authors of the original Semantic Web paper. Juan Sequeda is one of the people who has spent the last 2 decades bringing the Semantic Web principles, standards and technology to life in the real world. In this welcome address of the Connected Data London's 2026 Community Dinner, they put things into perspective, using both facts and a fun sketch. It's funny because it's true. All talk recordings from the event are being processed and will be available in the next few days. With so many amazing ideas and speakers at #CDL26, nobody could possibly catch them all in real-time. We've got you covered. ✅ If you were there, your ticket already includes replays. 🎟 If you did not make it, you can still get your remote ticket for perpetual on-demand access. 👉 sbee.link/dup3wm9c6f Recorded live at Connected Data London 2026 | 11–12 November | Leonardo Royal Hotel London Tower Bridge #CDL26 #KnowledgeGraph #GraphRAG #Ontology #Graph #AI #DataScience #GraphDB #SemTech

8:55
The Year of the Graph avatar
⟳ Connected Data retweeted
The Year of the Graph@TheYotg·Nov 12

Knowledge Graph Lakehouse: A Cloud-Native Architecture for Managing Enterprise Knowledge Graphs at Scale What if the knowledge graph were built only when a question asks for it? Enterprise knowledge graphs model connected domains well. At scale, volume, velocity and variety turn a single monolithic graph into something slow to ingest, hard to govern and often out of date. The Knowledge Graph Lakehouse inverts the order: Source documents land unchanged in a data lake A contextual index records which contexts each document is valid for (time, location, topic) Application-specific contextualized KGs are constructed on demand, only for the contexts a query names Interpretation is deferred to query time. What the system holds durably is the documents and the index. The design is grounded in KG-OLAP: contexts are cube cells, slice, dice and roll-up select and merge them, and knowledge propagates from covering contexts. Results are delivered as RDF quads, a labeled property graph or a GraphFrame. 🌍 Environmental data shows what this looks like in practice Regulated facilities change ID numbers, sit under parent corporations that acquire and spin off sites, and trade waste with sister facilities across overlapping jurisdictions. The records are spread over EPA systems such as RCRAInfo, ECHO and TRI, plus SEC filings. One lakehouse, three graphs: PFAS in a watershed: facilities, permits, pollutants, receiving waters, corporate parents Enforcement: violations, inspections, consent decrees, responsible legal entities Waste tracking: generators, transporters, manifest chains The lakehouse stays authoritative. The graph becomes a disposable perspective. 📊 Evaluation (air traffic management): Corpus grew from 28,590 to 324,456 documents; construction time for a fixed scope stayed near 2 seconds The index stayed around 1% of corpus size A new batch of 255 documents became queryable in about 18 seconds, against a 55-minute rebuild of a materialized cube Peak memory: about 1.4 GB on demand, 46 GB for the materialized build The architecture suits workloads whose scope is declared before any data is touched. Arbitrary queries over a standing graph remain better served by a materialized store. By B. Ahmad, C. G. Schuetz and L. Bozzato. Example by David G. Smith. davidgsmith.net/thoughts/why… #DataLakehouse #CloudNative #KGOLAP #DataArchitecture #ContextualKnowledgeGraphs -- 📩 The Year of the Graph Autumn 2026 newsletter issue is out! Meaning, Traversal, Reasoning, Action: The Graph Stack for Agentic AI. 👇 yearofthegraph.xyz/newslette… All things #KnowledgeGraph, #GraphDB, Graph #Analytics / #DataScience / #AI and #SemTech. Subscribe and follow to be in the know. Reach out if you’d like to be featured

Post attachment
The Year of the Graph avatar
⟳ Connected Data retweeted
The Year of the Graph@TheYotg·Nov 11

Which Local Model Should Build Your Knowledge Graph? A Reproducible Leaderboard A local model that chats well can still build a messy knowledge graph. Most local-model leaderboards score chat. A knowledge graph gives a model a different job first: read a passage, name the entities and relationships in a strict format, and point to the sentence that proves each one. Only later does it answer a question from what was extracted, without inventing anything. Those two jobs rank models differently. This leaderboard scores 22 local Ollama models on both, using pass/fail checks and no judge model. 🔎 What is measured: - Extraction: 65 isolated probes, plus 39 spliced into real chunks of War and Peace, covering output format, evidence, entity and relationship discipline - Chat: identical questions on Book One, answered from the same reference graph and retrieved context. An answer passes only if it names what was asked, nothing the context lacks, and declines when the book is silent - Overall: 0.6 extraction and 0.4 chat, with in-chunks extraction counted twice What the board shows: - The chat column is bunched. The extraction column spreads wide - Size is a weak predictor. Gemma 4 12B sits among models needing far more GPU memory, while several 30B-class models rank low as their in-chunks extraction falls apart - Gemma 4 31B leads on Overall and also built the reference graph behind the chat column - Every score carries a 90% interval from 2,000 resamples, and neighbouring rows are often ties - Flags explain low scores: answers cut off at the output limit, runaway repetition, or reasoning despite thinking being switched off - Models without Ollama tool calling can be scored but cannot serve as the chat model in the app The board is reproducible with a handful of CLI commands (pipx install chaoscypher-cli, then run the probes and chat-probes benchmarks). Exact numbers shift with Ollama versions and quantizations. The clusters should hold. The scores reflect each model together with ChaosCypher’s prompts and pipeline, on one novel for the chat column. The result is a basis for shortlisting models, not for settling on one. By Denis MacPherson dev.to/chaoscypherinc/which-… #GraphRAG #Ollama #LocalLLM #SelfHosted #Benchmarks -- 📩 The Year of the Graph Autumn 2026 newsletter issue is out! Meaning, Traversal, Reasoning, Action: The Graph Stack for Agentic AI. 👇 yearofthegraph.xyz/newslette… All things #KnowledgeGraph, #GraphDB, Graph #Analytics / #DataScience / #AI and #SemTech. Subscribe and follow to be in the know. Reach out if you’d like to be featured

Post attachment
The Year of the Graph avatar
⟳ Connected Data retweeted
The Year of the Graph@TheYotg·Nov 10

🦋 Catch the Butterfly: A Series A, and the Graph Engines continent behind it Why is a sales target being missed? An agent rules out region, sales team, pricing and competitors, one by one. What remains: slipped deals that share products with long delivery delays. Deals → products → components → a supplier behind on one part. Three hops from the original data. Nowhere a person would think to start. That investigation anchors PuppyGraph’s $10M Series A (led by Particle Future, with defy.vc, $15M raised in total).  The ideas underneath travel beyond one company: - Copilots answer questions a person already framed. Agents plan, query, read the result, and let it choose the next query. - Text-to-SQL works when relationships are known. Across thousands of tables, two integer ID columns (student, employee) join cleanly and mean nothing. - A failed query is visible. A successful query on the wrong relationship is not, and the next question inherits it. - An ontology in the query path can reject undefined entities and return structured feedback. People approve relationships, not every question. - Testing many hypotheses is query-hungry, so latency caps what an agent can explore. Discovery needs broad access, so permissions have to hold at each traversal step. The announcement also describes a category: an engine running multi-hop traversals over existing databases, warehouses and lakes, with no separate graph copy. State of the Graph defines Graph Engines as the layer that computes over existing systems, with no primary graph storage of their own. In the November v2.0 map (212 listings, 172 products, 146 vendors), the Graph Engines continent holds 7. All query data where it lives. None is a system of record. Four run on Snowflake or cloud warehouses. Zero-ETL is one of two forces reshaping the market, beside agentic AI and GraphRAG. Catch the Butterfly: PuppyGraph Raises $10M Series A to Build the Data Engine for AI Agents puppygraph.com/blog/series-a Surveying the Graph Engines Terrain stateofthegraph.com/2026/05/… 7 Continents, 5 Shifts: The Graph Technology Map, Redrawn stateofthegraph.com/2026/11/… #GraphEngines #AIAgents #ZeroETL #Ontology #Lakehouse -- 📩 The Year of the Graph Autumn 2026 newsletter issue is out! Meaning, Traversal, Reasoning, Action: The Graph Stack for Agentic AI. 👇 yearofthegraph.xyz/newslette… All things #KnowledgeGraph, #GraphDB, Graph #Analytics / #DataScience / #AI and #SemTech. Subscribe and follow to be in the know. Reach out if you’d like to be featured

Post attachment
Connected Data avatar
Connected Data@Connected_Data·Nov 10

Tomorrow, one of the authors of the original Semantic Web paper will speak at Connected Data London 2026 about what 25 years of linked data means for AI. You can be in the room at the Leonardo Royal Hotel, Tower Bridge. Over the next two days you'll also hear: 🎤 William Tunstall-Pedoe, founder of Evi (now part of Alexa), on breaking AI's trust ceiling 🎤 Joe Reis on why data engineers and knowledge graph people need each other 🎤 Juan Sequeda of ServiceNow with 20 lessons from 20 years of building ontologies 🎤 Tara Raafat (Bloomberg), Dagmar Beyer (Siemens) and Google's BigQuery Graph team on knowledge graphs and agents in production Plus, there are 12 hands-on masterclasses on Wednesday and a Community Dinner that evening. ⭐ Passes are still available, but not for long: lnkd.in/dSqXyC8T?utm_source=… #CDL26 #KnowledgeGraphs #AgenticAI #SemanticWeb #AI

Post attachment
The Year of the Graph avatar
⟳ Connected Data retweeted
The Year of the Graph@TheYotg·Nov 9

Graphs can help solve the AI agent memory problem Agents have a memory problem, Dave Bechberger notes: close the session, and everything they learned is gone. A useful long-term memory layer has to satisfy four properties: persistence, relevance retrieval, relational structure, and temporal validity. Vector stores nail the first two. But relationships and time are second-class citizens in a flat index; there’s no native way to ask “what led to what” or “what did we believe last month.” That’s the gap graphs close. Every edge should carry two independent clocks: valid time (when a fact was true in the world) and transaction time (when the system learned it). Nothing gets erased. Old edges get marked superseded as new evidence arrives, which makes “what did we believe last quarter” a query instead of a mystery. Left unmanaged, though, a memory graph rots: duplicate nodes, stale edges, quiet contradictions. Entity resolution, contradiction handling, decay, and compaction are what keep it usable rather than a junk drawer. Worth noting: this isn’t free. Construction cost, schema drift, and cases where a flat vector store is simply enough are real tradeoffs. William Lyon explores types of agent memory, and Animesh Kumar explores the first wave of agent memory systems – defining what memory is, exploring history and different implementations, and pointing out six cracks. As both Kumar and Bechberger point out, evaluation of long-term memory systems remains unsolved. No clean benchmark exists yet for multi-hop accuracy or temporal correctness. --      📩 Excerpt from The Year of the Graph Autumn 2026 newsletter      Read "# Meaning, Traversal, Reasoning, Action: The Graph Stack for Agentic AI" with more sections, references and attribution here 👇   yearofthegraph.xyz/newslette… All things #KnowledgeGraph, #GraphDB, Graph #Analytics / #DataScience / #AI and #SemTech.

Post attachment
Connected Data avatar
Connected Data@Connected_Data·Nov 10

Joining the Dots: Knowledge Graphs for Investigators and Analysts A detective’s wall has cards and red string. The cards matter. The string matters more. In most newsrooms and think tanks, the cards sit in spreadsheets, PDFs and timelines. The string lives in somebody’s head. The article reframes investigative work as a supply chain, and each knowledge graph technology maps onto a stage: 🔹 Parts: facts recorded as RDF triples (subject, verb, object), with each person, company and address given its own identifier. “Strings to things.” 🔹 Catalogue: an ontology works as a style guide for data. FollowTheMoney, Open Ownership and GLEIF already publish reusable ones. 🔹 Goods-in inspection: SHACL checks that every claim has a source and every date is coherent. 🔹 Assembly: Datalog rules trace control through layers of shell companies, as in the US Treasury’s 50 Percent Rule, and every conclusion can be traced back to its parts. 🔹 Bill of materials: named graphs and PROV record who said what, and when. The payoff is the manufacturer’s recall question turned round. If a source proves unreliable, which facts, arguments and published pieces relied on it? The limits are stated plainly too. A graph cannot tell whether a source is lying, and deciding that two records describe the same person is an editorial judgement. The suggested way in: write down the five questions you wish you could ask, reuse an existing vocabulary, and begin with one investigation and a few hundred statements. Every published story has a bill of materials, whether or not anyone writes it down. Knowledge graphs offer a rigorous way to write it down. 🔗 By Peter Winstanley sbee.link/uec6xvjf3y #SPARQL #RDF #SHACL #Datalog #InvestigativeJournalism -- Connected Data London 2026 | 11–12 November | Leonardo Royal Hotel London Tower Bridge 🎟 Tickets on sale now - 🔔 Full programme announced sbee.link/4jvebdmk96 📺 Sponsorship opportunities available. Contact info@connected-data.london for details. #KnowledgeGraph #GraphRAG #Ontology #Graph #AI #DataScience #GraphDB #SemTech

Post attachment
Connected Data avatar
Connected Data@Connected_Data·Nov 10

Norway's railway, Tripadvisor and TUI all depend on huge networks of data that never stop changing. This week at Connected Data London 2026, three people building knowledge graphs for them will show how they do it: The Travel Intent Graph: Joining Demand to Supply When One Experience Is Sold Many Times Over 🎤 Dennis Irorere, Tripadvisor The same tour or experience can turn up in many different listings. Dennis explains how a travel intent graph connects what people are searching for with what's actually available to book. 📅 Wed 11 Nov, 1:45 PM · Rapid Insights (Ludgate Suite) Build Your Own Enterprise Architecture Intelligence Pipeline with Ontology-Guided Conversational AI 🎤 Kalin Maldzhanski, TUI A hands-on masterclass for anyone who has to explain how a large organisation's systems fit together. Kalin walks through building a pipeline that answers enterprise architecture questions in plain language, with an ontology keeping the answers grounded. 📅 Wed 11 Nov, 3:30 PM · Masterclass (Warwick Suite) Building the Norwegian Railway Knowledge Graph: Where the Semantic Web Stack Runs Off the Track 🎤 Mathias Vanden Auweele, Matdata Mathias has been building a knowledge graph for Norway's railway. He'll be honest about where standard semantic web tools did the job, and where they fell short. 📅 Thu 12 Nov, 1:40 PM · Educational (Bartholomew Suite) Dennis and Kalin are both on Wednesday afternoon, so travel and transport teams can catch both in one afternoon! ⭐ The event starts tomorrow! Book your CDL26 pass and plan your sessions before it's too late: sbee.link/cjuar6dpye #CDL26 #KnowledgeGraphs #TravelTech #Rail #EnterpriseArchitecture #Ontology

Post attachmentPost attachmentPost attachmentPost attachment
Connected Data avatar
Connected Data@Connected_Data·Nov 9

Everyone is talking about agent memory and enterprise knowledge graphs, but very few people have noticed they’re turning into the same thing. At Connected Data London 2026, three sessions show what that means for anyone building AI agents: Context Graphs, Context Layers, and the Semantic Renaissance 🎤 Jessica Talisman, Ontology Pipeline Software engineers and data teams are both racing to give agents better context, from different directions. Jessica explains why they’re converging on knowledge graphs, and what each side can learn from the other. 📅 Thu 12 Nov, 11:00 AM · Edges (Salisbury Suite) When the LLM Wiki Meets Linked Data: Local Agent Memory That Scales 🎤 Tony Seale, The Knowledge Graph Guys A folder of markdown notes is the simplest agent memory there is, and builders love it, but it doesn’t scale. Tony shows how adding linked-data structure keeps it simple while making it work across an enterprise. 📅 Thu 12 Nov, 11:35 AM · Edges (Salisbury Suite) The Context Layer: From Semantics to Memory 🎤 Afraz Jaffri (Gartner), Atanas Kiryakov (Graphwise) & Mark Hahnel (Digital Science), moderated by Maya Natarajan What the context layer really contains, how trusted context reaches an agent, and what to ask any vendor who says they can deliver it. 📅 Thu 12 Nov, 3:50 PM · Nodes (Warwick Suite) The first two sessions run back-to-back in the Salisbury Suite on Thursday morning, so you can stay put for both! ⭐ 2 days to go! Book your pass and plan your sessions now: sbee.link/p8kd9vwqet #CDL26 #AgenticAI #AIagents #AgentMemory #KnowledgeGraphs #ContextLayer

Post attachmentPost attachmentPost attachmentPost attachment