The Ontologist: Michael Airhart

Custom graphs. Full control.

At #NODES2025 by @neo4j.com, Tony Smid explains why designing your own visualisation can outperform any plug-and-play tool.

📅 6 Nov, 13:00–13:30 CET
🔗 Learn more & register to join his session: neo4j.com/nodes-2025/a...

👀 Discover how to transform natural language questions into accurate #Cypher queries using an innovative loop of verification and correction—ensuring reliable results every time.

Explore this process! https://bit.ly/4pYpvlu


#Neo4j Makbule Gulcin Ozsoy

Google's New BlockRank Democratizes Advanced Semantic Search Google's BlockRank performed competitively with other state-of-the-art ranking models that were tested in research.

Google’s New BlockRank Democratizes Advanced Semantic Search Google's new BlockRank AI ranking method could make advanced semantic search accessible to everyone. The post Google’s New Block...

#Generative #AI #News #SEO

Origin | Interest | Match

Introducing Agentic Postgres, the Database for AI Agents Tiger is launching a new database for agents. Agentic Postgres is the first database designed from the ground up for Agents. It includes native full-text and semantic search built directly into the database. It has a new copy-on-write block storage layer that makes databases instantly forkable.

Introducing Agentic Postgres, the Database for AI Agents

Tiger is launching a new database for agents. Agentic Postgres is the first database designed from the ground up for Agents. It includes native full-text and semantic search built directly into the database. It has a new c…

#hackernews #news

Implementing DRIFT Search with Neo4j and LlamaIndex | Towards Data Science Combining global and local search to get the most accurate response

For those using @neo4j.com and @llamaindex.bsky.social, combining global and local search effectively can be a challenge. Tomaz Bratanic's new article demonstrates an implementation of DRIFT search to achieve more accurate and comprehensive responses from a knowledge graph.

NODES 2025 - Speaker Roundtable YouTube video by Neo4j

I am so hyped for #nodes25 in just two weeks! We kick things off with an amazing speaker roundtable! Tune in:
youtube.com/live/_DGziSs... #neo4j

Follow Bluesky's Semantic Web feed: knowledge graphs, linked data, and data taxonomy and ontology modeling.

Including Apache Jena, Cambridge Semantics Anzo, IBM Watson, PoolParty, Progress Semaphore, RDF, Stardog, Allegrograph, GraphDB, Graphwise, Neo4j, Ontotext, SPARQL, TopBraid and more.

Knowledge Graphs Gain Traction as AI Pushes Beyond Traditional Data Models Is graph really the new star schema? What do graphs like to non-insiders, and what attracts them to the community, methodologies, applications, and innovation?

Knowledge Graphs Gain Traction as AI Pushes Beyond Traditional Data Models

Is graph really the new star schema? What do graphs like to non-insiders, and what attracts them to the community, methodologies, applications, and innovation?

#hackernews #news

GitLab 18.4 released with GitLab Duo Model Selection and GitLab Knowledge Graph

about.gitlab.com/releases/...
GitLab 18.4 released with GitLab Duo Model Selection and GitLab Knowledge Graph

ORACLE: Ontology‑Driven Multi‑Hop Reasoning Boosts LLMs

ORACLE lets LLMs handle multi‑hop queries by auto‑building a knowledge graph, converting it to First‑Order Logic and answering sub‑questions, matching fine‑tuned scores without extra training. Read more: https://getnews.me/oracle-ontology-driven-multi-hop-reasoning-boosts-llms/ #oracle #llm

📦 scrypster/memento 0.9.2-beta

Advanced Neo4j knowledge graph memory system with MCP integration for AI tools and IDEs

🔗 https://bitbucket.org/scrypster/memento/

This makes my heart happy, and it’s nice that the blog can now afford a graph database to play with (graphrag, knowledge graphs, network science using graphdb). Never expected to see this, ever. Substack is digging my writing! #100daysofnetworks

Social SEO: Using Social to Build a Brand Knowledge Graph Learn how social media and secondary channels are essential for building a brand knowledge graph, which is key for ranking in AI-powered search.

To rank in AI-powered search, your brand must be a trusted entity within the knowledge graph.
Here's more mafostmarketing.com/social-brand...

You often hear that IKEA is a prominent user of knowledge graphs... But how exactly?

At #SEMANTiCS2025, Christelle Maignan, the lead ontologist @IKEA, lifts the curtain on the #IKEAKnowledgeGraph & talks real-world KG implementation at scale.

#knowledgegraphs #semantics #ikea

Liberation of data: how to liberate data from publications and how to reuse it #### **Description** This training session will offer participants a deep dive into the processes of annotating and sharing biodiversity data. The session will begin with an online theoretical introduction, providing an overview of the current state of biodiversity literature, the move towards the liberation of taxonomic data and the new opportunities this offers. This will be followed by two days of hands-on, in-person training focused on practical workflows for structuring, annotating, and circulating biodiversity data from scientific publications. Participants will explore how to annotate biodiversity datasets, ensuring they are properly formatted for reuse, and how to facilitate the sharing and circulation of these datasets within the scientific community. Trainees will also learn how to access and reuse data for their own research needs. By the end of the session, attendees will be equipped with the skills to enhance the accessibility and reusability of biodiversity data, contributing to more effective data exchange. ### **Online session** **15 October 2025** **14.00-16.00 Introduction: The state of biodiversity literature and the move towards data liberation** * Welcome: getting to know each other and trainees’ expectations * General introduction to the course * Technical requirements ### **Face to face** **10 November 2025** **Day 1: Structuring and annotating biodiversity data** **09.00-10.00 Module 1: Concepts** * Principles of semantics * The elements structuring a biodiversity publication * Annotations, attributes and linking: FAIR data * Workflows: template, individual extraction, born digital _versus_ scanned documents **10.00-10.30 – Coffee break** **10.30-13.00 Module 2: XML-first workflow** * Introduction to the XML-first workflow * Structuring a taxonomy/biodiversity paper * XML conversion and enrichment of the metadata * Annotating the data in JATS-Taxpub **13.00-14.00 – Lunch** **14.00-15.30 Module 3: Golden Gate** * The PDF workflow to enable re-use of data * Conversion and annotation of legacy literature * Learning and becoming a certified contributor **15.30-16.00 – Coffee break** **16.00-17.00 Module 3: Continuation** #### **11 November 2025** **Day 2:****Accessing and re-using biodiversity data** **9.00-10.30 Module 4: Curation and re-use of data from publications** * Curation and quality control of data already available in online repositories * TreatmentBank * Biodiversity Literature Repository * Synospecies * Ocellus **10.30-11.00 – Coffee break** **11.00-12.30 Module 4: Continuation** **12.30-13.30 – Lunch** **13.30-15.00 Module 5: Complementary re-use of data from publications** * Biodiversity PMC: specific annotation to question and answering * Biodiversity front end user interfaces * SIBiLS Collections * SIBiLS back-end services * SIBILS API & Data access channels * Curation-support tools * Advanced triage systems **15.00-15.30 – Coffee break** **15.30-17.00 Module 5: Continuation** * GBIF: occurrences to taxonomic names * GBIF – hosted portals * ChecklistBank: integrating names in CoL #### #### **Trainers** | **Laurence Bénichou** (https://orcid.org/0000-0002-0713-0751) is a French publisher. She is the Head of the Paris Museum Science Press (MNHN), and founded in 2011 the _European Journal of Taxonomy_ with a board of European colleagues and serves now as the Liaison officer for the journal. Since 2018, she leads the E-Publishing working group of CETAF. An expert in the field of scientific publishing for the Ministry of French higher education and research, she specialized in Linnaean Taxonomy publishing, Communication Design and Media. Her research is focused on diamond open access, digital publishing and data mining. ---|--- | **Chris Le Coquet** (https://orcid.org/0009-0006-7416-8983) is a French publisher. He is a desk editor for the _European Journal of Taxonomy_ , and is also in charge of digital projects focused on the FAIRisation of biodiversity data at the Paris Museum Science Press (MNHN). | **Donat Agosti** (https://orcid.org/0000-0001-9286-1200) is a Swiss biologist with over 30 years of experience, focused on making biodiversity data openly accessible. He co-founded Plazi in 2008, a Swiss NGO that develops workflows to convert scientific literature into FAIR data. Through partnerships with institutions like GBIF, NIH, Zenodo, and Biodiversity PMC, Plazi’s work enables the reuse of published data in global research infrastructures, with Plazi being the largest data contributor to Zenodo, GBIF and COL. He is also a widely published researcher in taxonomy and biodiversity informatics. | **Julia Giora** (https://orcid.org/0009-0006-7416-8983) is a Brazilian biologist with PhD and postdoctoral training in Animal Biology, with over 20 years of experience in research, higher education, and biodiversity data. Currently leading the Learning & Engagement team at Plazi, working internationally with FAIR scientific data and training. Also active as a content developer for universities and Brazilian NGOs focused on biodiversity conservation. Author of more than 20 scientific publications, including peer-reviewed articles, books, and book chapters. | **Emilie Pasche** (https://orcid.org/0000-0002-9118-5762) is a research associate at HES-SO Geneva and SIB, and is involved in Biodiversity PMC. | **Markus Döring** (https://orcid.org/0000-0001-7757-1889) is a German botanist and biodiversity informatician currently working for the Global Biodiversity Information Facility (GBIF) in Copenhagen, Denmark. Trained originally as a botanist, Markus bridges the gap between taxonomy and data infrastructure. He has played a key role in developing the GBIF Backbone Taxonomy, ChecklistBank and the integrated publishing toolkit (IPT). He is the lead developer for the Catalogue of Life and has been engaged in Biodiversity Information Standards (TDWG) since 2001 with contributions to Darwin Core and TCS. **** #### **Dates of Training period** One theoretical online session, plus two working days (October to November 2025) divided as follows: Online session Wednesday 15th of October 2025 Face to face practical experience Monday 10th and Tuesday 11th of November 2025 #### #### **Location** Villa Engler – Freie Universität Berlin, Altensteinstr. 2, 14195 Berlin #### #### **Course’s language** English #### #### **Target audience** Editors / Publishers / Librarians / Researchers / Students #### #### **Fee** The school enrolment is free. All the other costs are at the expense of participants. #### **Registration deadline** 30 September 2025 #### #### **Mode of trainees’ assessment** * A short quiz will be provided at the end of the theoretical introduction; * Short exercises will be provided at the end of each practical module. #### #### **Participant quota (min and max number of trainees)** 10–20 (1-2 groups) #### #### **Types of training/ Implementation method** 1. Theoretical modules/Online and in-person lectures 2. Practical experience / face to face * Structuring and annotating biodiversity data. / Hands-on exercises using XML-first workflow and Golden Gate * Exploring and reusing biodiversity data / Hands-on exercises using TreatmentBank, BLR, GBIF, BiodiversityPMC, etc. #### #### **Training Course** **learning outcomes** The present course will cover a variety of topics from purely theoretical to the development of practical skills in the field of biodiversity data. The main expected outcomes are: * Learn about the state of biodiversity literature and the move to liberate biodiversity data from publications * Familiarise with the standards and workflows in biodiversity data * Practice the structuration of biodiversity paper * Practice the annotation of biodiversity data * Learn how to efficiently access and reuse biodiversity data * Practice data infrastructures such as TreatmentBank, GBIF and BiodiversityPMC * First step to become a certified contributor to TreatmentBank #### #### **Certifications provided** 1. Certificate of Attendance by CETAF DEST with 5 ECVET Units (_European Credit system for Vocational Education and Training_) 2. Certificate by CETAF DEST according to Europass Certificate Supplement (certifying analytically the knowledge, skills and competences gained) #### #### **What trainees need to bring** 1. Laptop (up-to-date OS) 2. Microsoft Word 3. Java ver. 23 4. Libre Office ver. 7.6.5.2 (provided) 5. XML Mind ver. 9.5.1 (provided) 6. Golden Gate Imagine (provided) **Registration form** **More details:**dest@cetaf.org

Liberation of data: how to liberate data from publications and how to reuse it - DEST (Distributed European School of Taxonomy) https://cetaf.org/dest/liberation-of-data-how-to-liberate-data-from-publications-and-how-to-reuse-it/

Survey of LLM Ensemble Methods for Combining Large Language Models

A new survey maps LLM ensemble methods, defining a three‑stage taxonomy (before‑, during‑, after‑inference) and benchmarks; it was updated in September 2025. Read more: https://getnews.me/survey-of-llm-ensemble-methods-for-combining-large-language-models/ #llmensemble #taxonomy

By the way, what’s a AA? Instead of fuzzy labels like 'indie' and 'AA', we propose a new data-driven system to classify video games based on their actual production scope. This new framework reveals the real economic and crea...

A taxonomy of the video game production scope

#gamedev #gamedevelopment #indiedev #gamedesign

I Thought I Knew Silicon Valley. I Was Wrong Tech got what it wanted by electing Trump. A year later, it looks more like a suicide pact.

A snapshot of Silicon Valley in 2025 is Mark Zuckerberg, now a MAGA-friendly mixed martial arts fan who doesn’t worry much about hate speech on his platforms and complains that corporate America isn’t masculine enough.

MediaWiki Users and Developers Conference Fall 2025 - semantic-mediawiki.org MediaWiki Users and Developers Conference Fall 2025. From semantic-mediawiki.org

📢 Contribute to the MediaWiki Users and Developers Conference Fall 2025!

🌍 Come together with developers, users, & organizations from the MediaWiki community for presentations & workshops/demos, & submit your ideas! 💡

📅 Submission deadline: September 5

www.semantic-mediawiki.org/wiki/MediaWi...

YouTube video by The Knowledge Graph Conference Knowledge Bites with Christos Giotis

"We're pushing the absolute limits of what can be done with graph hardware."
–Christos Giotis

In this week's Knowledge Bite video, Christos Giotis from the University of Edinburgh shares his work on Atlas, a hardware engine for #graphdatabases:

www.youtube.com/shorts/MTp5Z...

#knowledgegraphs

updated to include confirmation, Semantic Search in web of Science smart search is using vector embedding webofscience.zendesk.com/hc/en-us/art....

Web authors were already using schema.org and other signals, but how much of the current crop of AI slop-slingers are doing so? And how much weight are semantic web tools being given in search ranking now?

Amazon Neptune Analytics now introduces stop/start capability Today, we are excited to announce support for Stop/Start in Amazon Neptune Analytics, a new capability that enables organizations to pause and resume their graph workloads on demand,helping reduce costs during idle periods without losing data or configuration. Many customers use Neptune Analytics for periodic graph workloads such as fraud detection, recommendation engines, or research simulations that run periodically. Until now, customers had to choose between keeping their Neptune Analytics graphs online even when not in use or deleting and recreating them each time they were needed. This approach was not only expensive, but also time-consuming, requiring manual infrastructure management, repeated data imports, and updates to downstream pipelines to accommodate each newly created graph. This adds significant operational overhead and complexity to their analytics workflows. With Stop/Start, customers can now pause a graph workload via the AWS Console, CLI, or API, and resume it later with a single action. While the graph is stopped, they pay only 10% of the normal compute cost, and all data and settings are preserved without needing to delete or rebuild graphs. This feature is particularly valuable for cost-conscious startups, research teams, and enterprises with analytics workloads. It simplifies lifecycle management and unlocks experimentation at lower price points. Stop/Start for Neptune Analytics is available in all commercial regions where Neptune Analytics is offered. You can start using this feature today via the Neptune Analytics console, AWS CLI, or AWS SDKs. To learn more, visit the documentation and the pricing page.

🆕 Amazon Neptune Analytics now offers Stop/Start, cutting costs by 90% during idle periods. Pause and resume workloads, preserving data and settings, simplifying management and reducing overhead. Available globally.

#AWS #AmazonNeptune

The missing piece for autonomous AI agents Subscribe • Previous Issues

Is GraphRAG Dead or Is It Just Evolving into Something Smarter?
🆕 developer-friendly tools like the Kùzu graph database are making it easier to implement graph-based systems, potentially expanding adoption
gradientflow.substack.com/p/the-missin...

🚨 New course with ‪@datacamp.bsky.social‬!

Discover how GraphRAG boosts the accuracy and reliability of RAG apps by storing and querying info as nodes and relationships.

🚀 Smarter RAG combining #Neo4j and LangChain 👨‍💻 with Adam Cowley.

https://bit.ly/4lVhsTo

#Langchain #RAG

A powerful use case for AI grounded by a knowledge graph (GraphRAG): Know-Your-Customer (KYC) - a critical defense line against illicit activities.

Find out how AI + knowledge graphs detect sophisticated financial crimes in KYC investigations.

https://bit.ly/46aSgDw

Agentic Knowledge Graph Construction Build a multi-agent system that plans, designs, and constructs a knowledge graph.

Agentic knowledge graph construction landed this week, and I still wake up thinking about improvements I want to make to it.

www.deeplearning.ai/short-course...

Achieving the Perfect Golden Record with Graph Data | Aerospike Learn how to achieve a perfect golden record for entity resolution using graph data models, AWS, and Aerospike for enhanced ad targeting and personalization.

The combination of #AWS entity resolution and #Aerospike Graph makes creating a modern golden record for entity and identity resolution use cases more efficient. Learn the challenges and solutions for entity resolution for #AdTech companies.

https://monkeylink.co/fba75e #Database #GraphDB

Microsoft launches inquiry into claims Israel used its tech for mass surveillance of Palestinians Company says use of its cloud technology to store millions of intercepted calls would breach terms of service Microsoft has launched an “urgent” external inquiry into allegations Israel’s military surveillance agency has used the company’s technology to facilitate the mass surveillance of Palestinians. The company said on Friday the formal review was in response to a Guardian investigation that revealed how the Unit 8200 spy agency has relied on Microsoft’s Azure cloud platform to store a vast collection of everyday Palestinian mobile phone calls. Continue reading...

Microsoft launches inquiry into claims Israel used its tech for mass surveillance of Palestinians

AI Eroded Doctors’ Ability to Spot Cancer Within Months in Study Artificial intelligence, touted for its potential to transform medicine, led to some doctors losing skills after just a few months in a new study.

Stop offloading cognitive tasks to "generative AI." Stop using systems w/ "AI" features either inextricably woven through them or prominently displayed at the top to nudge you into their use. Stop *Designing* "AI" tools & integrations that way. Stop building or using "AI" like this. Fucking Stop it.