home.social

#knowledgegraphs — Public Fediverse posts

Live and recent posts from across the Fediverse tagged #knowledgegraphs, aggregated by home.social.

fetched live
  1. 📢 Submit your research to swat4hcls.org by 14.Sep.2026 at app.oxfordabstracts.com/stages

    Accepted papers will have the chance to submit an extended version to the conference collection at the Journal of Biomedical Semantics

    Topics around for and

  2. Imagine if, instead of building #knowledgegraphs mapping the relationships and semantic meaning behind raw data ahead of time, networks of #AIagents could build their own, instantly. That's what DFT Labs, a new venture of a company called HeyDonto, proposes, along with a new set of benchmarks that gauge not just how well #AI models predict the next token, but how well they actually understand and preserve data context. HeyDonto's co-founder, CTO and chief researcher sits down with IT Ops Query to discuss his work, and its industry-wide implications.

    In today’s episode, we’ll cover…

    ·       Data Field Theory, a new approach to how AI systems learn

    ·       Potential AI compliance implications with self-updating systems

    ·       How quantum physics can inform data intelligence

    And more!

    Check it out here:

    youtu.be/c9DK33BAZ3s

  3. Imagine if, instead of building #knowledgegraphs mapping the relationships and semantic meaning behind raw data ahead of time, networks of #AIagents could build their own, instantly. That's what DFT Labs, a new venture of a company called HeyDonto, proposes, along with a new set of benchmarks that gauge not just how well #AI models predict the next token, but how well they actually understand and preserve data context. HeyDonto's co-founder, CTO and chief researcher sits down with IT Ops Query to discuss his work, and its industry-wide implications.

    In today’s episode, we’ll cover…

    ·       Data Field Theory, a new approach to how AI systems learn

    ·       Potential AI compliance implications with self-updating systems

    ·       How quantum physics can inform data intelligence

    And more!

    Check it out here:

    youtu.be/c9DK33BAZ3s

  4. Imagine if, instead of building #knowledgegraphs mapping the relationships and semantic meaning behind raw data ahead of time, networks of #AIagents could build their own, instantly. That's what DFT Labs, a new venture of a company called HeyDonto, proposes, along with a new set of benchmarks that gauge not just how well #AI models predict the next token, but how well they actually understand and preserve data context. HeyDonto's co-founder, CTO and chief researcher sits down with IT Ops Query to discuss his work, and its industry-wide implications.

    In today’s episode, we’ll cover…

    ·       Data Field Theory, a new approach to how AI systems learn

    ·       Potential AI compliance implications with self-updating systems

    ·       How quantum physics can inform data intelligence

    And more!

    Check it out here:

    youtu.be/c9DK33BAZ3s

  5. We are very happy about the publication of
    Semantic Modeling in Materials Science and Engineering With Platform MaterialDigital Core Ontology 3.0 in collaboration with
    Markus Schilling, Philipp von Hartrott, Jörg Waitelonis, Thomas Hanke, Henk Birkholz, Hossein Beygi Nasrabadi, Khashayar Razghandi, Kamilla Zaripova, Felix Thonagel, Fabian Neuhaus, Martin Glauer, Lars Vogt, @lysander07, Lutz Mädler, Bernd Bayerlein, Chris Eberl

    ‘ontologies #knowledgegraphs #mse #materialsscience @fiz_karlsruhe

  6. We are very happy about the publication of
    Semantic Modeling in Materials Science and Engineering With Platform MaterialDigital Core Ontology 3.0 in collaboration with
    Markus Schilling, Philipp von Hartrott, Jörg Waitelonis, Thomas Hanke, Henk Birkholz, Hossein Beygi Nasrabadi, Khashayar Razghandi, Kamilla Zaripova, Felix Thonagel, Fabian Neuhaus, Martin Glauer, Lars Vogt, @lysander07, Lutz Mädler, Bernd Bayerlein, Chris Eberl

    ‘ontologies #knowledgegraphs #mse #materialsscience @fiz_karlsruhe

  7. We are very happy about the publication of
    Semantic Modeling in Materials Science and Engineering With Platform MaterialDigital Core Ontology 3.0 in collaboration with
    Markus Schilling, Philipp von Hartrott, Jörg Waitelonis, Thomas Hanke, Henk Birkholz, Hossein Beygi Nasrabadi, Khashayar Razghandi, Kamilla Zaripova, Felix Thonagel, Fabian Neuhaus, Martin Glauer, Lars Vogt, @lysander07, Lutz Mädler, Bernd Bayerlein, Chris Eberl

    ‘ontologies #knowledgegraphs #mse #materialsscience @fiz_karlsruhe

  8. 🤠 #KDAI2026 final lecture on #KnowledgeGraphs 04 & #NeurosymbolicAI

    The final graph session, then the leap to hybrid AI:

    ▸ OWL — complex classes & property restrictions
    ▸ SHACL Shapes
    ▸ Neurosymbolic AI with KG embeddings, RAG,
    ▸ AI limits revisited (ELIZA, Clever Hans, Chinese Room, Turing Test, the paperclip maximiser & the singularity).

    Neither symbols nor learning alone get us there. 📡 SEE YOU SPACE COWBOY…

    #SemanticWeb #OWL #SHACL #RAG #Ontologies #LLM @fiz_karlsruhe @KIT_Karlsruhe #AI

  9. 🤠 #KDAI2026 final lecture on #KnowledgeGraphs 04 & #NeurosymbolicAI

    The final graph session, then the leap to hybrid AI:

    ▸ OWL — complex classes & property restrictions
    ▸ SHACL Shapes
    ▸ Neurosymbolic AI with KG embeddings, RAG,
    ▸ AI limits revisited (ELIZA, Clever Hans, Chinese Room, Turing Test, the paperclip maximiser & the singularity).

    Neither symbols nor learning alone get us there. 📡 SEE YOU SPACE COWBOY…

    #SemanticWeb #OWL #SHACL #RAG #Ontologies #LLM @fiz_karlsruhe @KIT_Karlsruhe #AI

  10. 🤠 #KDAI2026 final lecture on #KnowledgeGraphs 04 & #NeurosymbolicAI

    The final graph session, then the leap to hybrid AI:

    ▸ OWL — complex classes & property restrictions
    ▸ SHACL Shapes
    ▸ Neurosymbolic AI with KG embeddings, RAG,
    ▸ AI limits revisited (ELIZA, Clever Hans, Chinese Room, Turing Test, the paperclip maximiser & the singularity).

    Neither symbols nor learning alone get us there. 📡 SEE YOU SPACE COWBOY…

    #SemanticWeb #OWL #SHACL #RAG #Ontologies #LLM @fiz_karlsruhe @KIT_Karlsruhe #AI

  11. @IBI_HU … Forschungsdaten digitaler Editionen semantisch angereichert, geöffnet und nachnutzbar gemacht werden können, um ihre Überführung in Knowledge Graphs zu ermöglichen. Dabei werden unter anderem Gender Biases und epistemische ‚Lücken‘ in Wissensbasen, Fragen der Datenqualität und -herkunft sowie die Bedeutung Wissenschaftlicher Bibliotheken und Informationsinfrastrukturen diskutiert."
    #DigitaleEditionen #Forschungsdaten #KnowledgeGraphs

  12. @IBI_HU … Forschungsdaten digitaler Editionen semantisch angereichert, geöffnet und nachnutzbar gemacht werden können, um ihre Überführung in Knowledge Graphs zu ermöglichen. Dabei werden unter anderem Gender Biases und epistemische ‚Lücken‘ in Wissensbasen, Fragen der Datenqualität und -herkunft sowie die Bedeutung Wissenschaftlicher Bibliotheken und Informationsinfrastrukturen diskutiert."
    #DigitaleEditionen #Forschungsdaten #KnowledgeGraphs

  13. @IBI_HU … Forschungsdaten digitaler Editionen semantisch angereichert, geöffnet und nachnutzbar gemacht werden können, um ihre Überführung in Knowledge Graphs zu ermöglichen. Dabei werden unter anderem Gender Biases und epistemische ‚Lücken‘ in Wissensbasen, Fragen der Datenqualität und -herkunft sowie die Bedeutung Wissenschaftlicher Bibliotheken und Informationsinfrastrukturen diskutiert."
    #DigitaleEditionen #Forschungsdaten #KnowledgeGraphs

  14. 🕸️ Can AI make knowledge graphs usable for the people who actually need them?

    Gerald Hiebel showed how AI lets non-technical domain experts query knowledge graphs directly, without an ontology specialist composing or checking the query first. Skip that check and the expert has to trust the AI's answer on its own, raising real questions about explainability and reproducibility.

    Where would you use it, and where would you not trust it?

    #AI #knowledgegraphs #research

  15. 🕸️ Can AI make knowledge graphs usable for the people who actually need them?

    Gerald Hiebel showed how AI lets non-technical domain experts query knowledge graphs directly, without an ontology specialist composing or checking the query first. Skip that check and the expert has to trust the AI's answer on its own, raising real questions about explainability and reproducibility.

    Where would you use it, and where would you not trust it?

    #AI #knowledgegraphs #research

  16. 🕸️ Can AI make knowledge graphs usable for the people who actually need them?

    Gerald Hiebel showed how AI lets non-technical domain experts query knowledge graphs directly, without an ontology specialist composing or checking the query first. Skip that check and the expert has to trust the AI's answer on its own, raising real questions about explainability and reproducibility.

    Where would you use it, and where would you not trust it?

    #AI #knowledgegraphs #research

  17. #KDAI2026 lecture no 11. This time it's interrogation tactics: FILTER, REGEX, OPTIONAL, UNION, negation, BIND, and GROUP BY aggregates in SPARQL.

    Then a bounty comparison: DBpedia (1.32B triples, built from Wikipedia infoboxes) vs Wikidata (17.6B triples, ~29K editors)

    Closing out with OWL: Description Logic SROIQ(D), etc.

    See you space cowboy…

    #SemanticWeb #KnowledgeGraphs #SPARQL #DBpedia #Wikidata #OWL #LinkedData #AIeducation @fiz_karlsruhe @KIT_Karlsruhe @fizise #AI #cowboybebop

  18. #KDAI2026 lecture no 11. This time it's interrogation tactics: FILTER, REGEX, OPTIONAL, UNION, negation, BIND, and GROUP BY aggregates in SPARQL.

    Then a bounty comparison: DBpedia (1.32B triples, built from Wikipedia infoboxes) vs Wikidata (17.6B triples, ~29K editors)

    Closing out with OWL: Description Logic SROIQ(D), etc.

    See you space cowboy…

    #SemanticWeb #KnowledgeGraphs #SPARQL #DBpedia #Wikidata #OWL #LinkedData #AIeducation @fiz_karlsruhe @KIT_Karlsruhe @fizise #AI #cowboybebop

  19. #KDAI2026 lecture no 11. This time it's interrogation tactics: FILTER, REGEX, OPTIONAL, UNION, negation, BIND, and GROUP BY aggregates in SPARQL.

    Then a bounty comparison: DBpedia (1.32B triples, built from Wikipedia infoboxes) vs Wikidata (17.6B triples, ~29K editors)

    Closing out with OWL: Description Logic SROIQ(D), etc.

    See you space cowboy…

    #SemanticWeb #KnowledgeGraphs #SPARQL #DBpedia #Wikidata #OWL #LinkedData #AIeducation @fiz_karlsruhe @KIT_Karlsruhe @fizise #AI #cowboybebop

  20. With a career that spans academic study as a computational neuroscientist and his current work creating enterprise systems that ground LLMs in ontology-backed #knowledgeGraphs, Yann Le Franc is a living embodiment of #neurosymbolicAI.

    He's also a super-experienced #ontology designer who helps other practitioners get out of their practice silos and adopt time-tested semantic modeling methods and tools.

    knowledgegraphinsights.com/yan

  21. With a career that spans academic study as a computational neuroscientist and his current work creating enterprise systems that ground LLMs in ontology-backed #knowledgeGraphs, Yann Le Franc is a living embodiment of #neurosymbolicAI.

    He's also a super-experienced #ontology designer who helps other practitioners get out of their practice silos and adopt time-tested semantic modeling methods and tools.

    knowledgegraphinsights.com/yan

  22. I'm hosting August's carnival! The topic is "The Search for Knowledge".

    Check out the post for the topic and how to submit.

    chiply.dev/post-august-emacs-c

  23. Back in LA after #DwebCamp in Alte Hölle; migrating now to #Fediverse 👋 please welcome my #introduction

    My middle name comes from 新 (new) 泳 (swimming); meaning? my work is to bridge cultures and open new ways of thinking; we've got to move some rocks to let the waters flow

    Trained as a documentary filmmaker, now designing multimodal systems for interdisciplinary studies ( #obsidian and #knowledgegraphs ); i'm currently catching up with #examprep for my PhD #studywithme #undercommons #autofiction #storytime

    SPOTTED: Did you know? Palantir's grounding ideologies actually derived from Frankfurt school pedagogy. How leftist cultural theory could become co-opted into a techno-fascist moral catastrophe signals at a limit to interpretive method. You'll always find me gravitating toward #reparative approaches to education, arts, urban science, technology, etc. and in dialogue across sectors #letstalk

  24. Back in LA after #DwebCamp in Alte Hölle; migrating now to #Fediverse 👋 please welcome my #introduction

    My middle name comes from 新 (new) 泳 (swimming); meaning? my work is to bridge cultures and open new ways of thinking; we've got to move some rocks to let the waters flow

    Trained as a documentary filmmaker, now designing multimodal systems for interdisciplinary studies ( #obsidian and #knowledgegraphs ); i'm currently catching up with #examprep for my PhD #studywithme #undercommons #autofiction #storytime

    SPOTTED: Did you know? Palantir's grounding ideologies actually derived from Frankfurt school pedagogy. How leftist cultural theory could become co-opted into a techno-fascist moral catastrophe signals at a limit to interpretive method. You'll always find me gravitating toward #reparative approaches to education, arts, urban science, technology, etc. and in dialogue across sectors #letstalk

  25. Back in LA after #DwebCamp in Alte Hölle; migrating now to #Fediverse 👋 please welcome my #introduction

    My middle name comes from 新 (new) 泳 (swimming); meaning? my work is to bridge cultures and open new ways of thinking; we've got to move some rocks to let the waters flow

    Trained as a documentary filmmaker, now designing multimodal systems for interdisciplinary studies ( #obsidian and #knowledgegraphs ); i'm currently catching up with #examprep for my PhD #studywithme #undercommons #autofiction #storytime

    SPOTTED: Did you know? Palantir's grounding ideologies actually derived from Frankfurt school pedagogy. How leftist cultural theory could become co-opted into a techno-fascist moral catastrophe signals at a limit to interpretive method. You'll always find me gravitating toward #reparative approaches to education, arts, urban science, technology, etc. and in dialogue across sectors #letstalk

  26. Back in LA after #DwebCamp in Alte Hölle; migrating now to #Fediverse 👋 please welcome my #introduction

    My middle name comes from 新 (new) 泳 (swimming); why? because my work is to bridge cultures and open ways of thinking;
    we've got to move some rocks to let the waters flow & just keep swimming~

    Trained as a documentarian, now designing multimodal systems for interdisciplinary studies ( #obsidian and #knowledgegraphs for now); catching up on my PhD duties #undercommons #examprep #studywithme

    SPOTTED: Did you know? Palantir's grounding ideologies derived from Frankfurt school pedagogy. How leftist cultural theory could become co-opted into a techno-fascist moral catastrophe is a question on method. Always toward #reparative approaches to education, arts, urban science, technology, etc. will be the corner you find me

  27. Back in LA after #DwebCamp in Alte Hölle; migrating now to #Fediverse 👋 please welcome my #introduction

    My middle name comes from 新 (new) 泳 (swimming); why? because my work is to bridge cultures and open ways of thinking;
    we've got to move some rocks to let the waters flow & just keep swimming~

    Trained as a documentarian, now designing multimodal systems for interdisciplinary studies ( #obsidian and #knowledgegraphs for now); catching up on my PhD duties #undercommons #examprep #studywithme

    SPOTTED: Did you know? Palantir's grounding ideologies derived from Frankfurt school pedagogy. How leftist cultural theory could become co-opted into a techno-fascist moral catastrophe is a question on method. Always toward #reparative approaches to education, arts, urban science, technology, etc. will be the corner you find me

  28. From raw statements to sound inference. This week's #KDAI2026 session works through the RDF stack top to bottom:
    ▸RDF — typed literals, blank nodes, and Turtle
    ▸RDFS — classes, domains & ranges, subClassOf hierarchies, model-theoretic semantics, reification & RDF*
    ▸SPARQL — first steps

    The triple is the message. 📡

    @fiz_karlsruhe @fizise @KIT_Karlsruhe #semanticweb #RDF #KnowledgeGraphs #LinkedData #AI #lecture #sparql

  29. From raw statements to sound inference. This week's #KDAI2026 session works through the RDF stack top to bottom:
    ▸RDF — typed literals, blank nodes, and Turtle
    ▸RDFS — classes, domains & ranges, subClassOf hierarchies, model-theoretic semantics, reification & RDF*
    ▸SPARQL — first steps

    The triple is the message. 📡

    @fiz_karlsruhe @fizise @KIT_Karlsruhe #semanticweb #RDF #KnowledgeGraphs #LinkedData #AI #lecture #sparql

  30. From raw statements to sound inference. This week's #KDAI2026 session works through the RDF stack top to bottom:
    ▸RDF — typed literals, blank nodes, and Turtle
    ▸RDFS — classes, domains & ranges, subClassOf hierarchies, model-theoretic semantics, reification & RDF*
    ▸SPARQL — first steps

    The triple is the message. 📡

    @fiz_karlsruhe @fizise @KIT_Karlsruhe #semanticweb #RDF #KnowledgeGraphs #LinkedData #AI #lecture #sparql

  31. 🌐 Join Our Team: Shape the Future of Neurosymbolic AI & Knowledge Graphs! 🚀
    We are looking for a PhD/Junior Researcher or PostDoc/Senior Researcher with research interest in neurosymbolic AI, knowledge graphs, ontologies, and large language models to work with us in the Platform MaterialDigital project.

    Find more details & apply at
    fiz-karlsruhe.de/en/stellenanz

    @fizise @fiz_karlsruhe @KIT_Karlsruhe #knowledgegraphs #AI #llms #ontologies #neurosymbolicAI #semanticweb #joboffer #PhDJobs #PostdocJobs

  32. 🌐 Join Our Team: Shape the Future of Neurosymbolic AI & Knowledge Graphs! 🚀
    We are looking for a PhD/Junior Researcher or PostDoc/Senior Researcher with research interest in neurosymbolic AI, knowledge graphs, ontologies, and large language models to work with us in the Platform MaterialDigital project.

    Find more details & apply at
    fiz-karlsruhe.de/en/stellenanz

    @fizise @fiz_karlsruhe @KIT_Karlsruhe #knowledgegraphs #AI #llms #ontologies #neurosymbolicAI #semanticweb #joboffer #PhDJobs #PostdocJobs

  33. 🌐 Join Our Team: Shape the Future of Neurosymbolic AI & Knowledge Graphs! 🚀
    We are looking for a PhD/Junior Researcher or PostDoc/Senior Researcher with research interest in neurosymbolic AI, knowledge graphs, ontologies, and large language models to work with us in the Platform MaterialDigital project.

    Find more details & apply at
    fiz-karlsruhe.de/en/stellenanz

    @fizise @fiz_karlsruhe @KIT_Karlsruhe #knowledgegraphs #AI #llms #ontologies #neurosymbolicAI #semanticweb #joboffer #PhDJobs #PostdocJobs

  34. The new #KDAI2026 lecture is out. How to get from a sentence to something a machine can actually reason?

    The greenhouse effect was discovered by Fourier, explained by Eunice Newton Foote. 3 words, 1 triple: subject, predicate, object. Stack enough of them and you get a graph. Add an ontology and the machine starts to infer.

    This session:
    ▸ Graphs & triples
    #KnowledgeGraphs & #ontologies
    #SemanticWeb & #LinkedData
    #RDF, RDFS, #SPARQL, #OWL, #SHACL
    CU space cowboy…

    @fiz_karlsruhe @fizise

  35. The new #KDAI2026 lecture is out. How to get from a sentence to something a machine can actually reason?

    The greenhouse effect was discovered by Fourier, explained by Eunice Newton Foote. 3 words, 1 triple: subject, predicate, object. Stack enough of them and you get a graph. Add an ontology and the machine starts to infer.

    This session:
    ▸ Graphs & triples
    #KnowledgeGraphs & #ontologies
    #SemanticWeb & #LinkedData
    #RDF, RDFS, #SPARQL, #OWL, #SHACL
    CU space cowboy…

    @fiz_karlsruhe @fizise

  36. The new #KDAI2026 lecture is out. How to get from a sentence to something a machine can actually reason?

    The greenhouse effect was discovered by Fourier, explained by Eunice Newton Foote. 3 words, 1 triple: subject, predicate, object. Stack enough of them and you get a graph. Add an ontology and the machine starts to infer.

    This session:
    ▸ Graphs & triples
    #KnowledgeGraphs & #ontologies
    #SemanticWeb & #LinkedData
    #RDF, RDFS, #SPARQL, #OWL, #SHACL
    CU space cowboy…

    @fiz_karlsruhe @fizise

  37. Engineering Interpretation: #DigitalIdentityOptimization as a unified theory of #digitalvisibility

    Entity understood as #StabilizedInterpretiveNode in #LatentSemanticSpace - not metaphysical “things,” but nodes of meaning interpreted across #AI, #search, and #knowledgegraphs

    #DIO/#ODI = meaning syndication + #semanticSEO + #knowledgegraphs + #LLMvisibility - united through invariant cycle: entity → representation → interpretation → trust → relationship → reconstruction

    slideshare.net/slideshow/engin

  38. Engineering Interpretation: #DigitalIdentityOptimization as a unified theory of #digitalvisibility

    Entity understood as #StabilizedInterpretiveNode in #LatentSemanticSpace - not metaphysical “things,” but nodes of meaning interpreted across #AI, #search, and #knowledgegraphs

    #DIO/#ODI = meaning syndication + #semanticSEO + #knowledgegraphs + #LLMvisibility - united through invariant cycle: entity → representation → interpretation → trust → relationship → reconstruction

    slideshare.net/slideshow/engin

  39. Engineering Interpretation: #DigitalIdentityOptimization as a unified theory of #digitalvisibility

    Entity understood as #StabilizedInterpretiveNode in #LatentSemanticSpace - not metaphysical “things,” but nodes of meaning interpreted across #AI, #search, and #knowledgegraphs

    #DIO/#ODI = meaning syndication + #semanticSEO + #knowledgegraphs + #LLMvisibility - united through invariant cycle: entity → representation → interpretation → trust → relationship → reconstruction

    slideshare.net/slideshow/engin

  40. Engineering Interpretation: #DigitalIdentityOptimization as a unified theory of #digitalvisibility

    Entity understood as #StabilizedInterpretiveNode in #LatentSemanticSpace - not metaphysical “things,” but nodes of meaning interpreted across #AI, #search, and #knowledgegraphs

    #DIO/#ODI = meaning syndication + #semanticSEO + #knowledgegraphs + #LLMvisibility - united through invariant cycle: entity → representation → interpretation → trust → relationship → reconstruction

    slideshare.net/slideshow/engin

  41. #RAG gave #LLMs access to external knowledge - but standard retrieval systems still struggle.

    #GraphRAG takes a different approach by using knowledge graphs to map entities, relationships, and provenance. The result?

    Watch the #InfoQ video by Cassie Shum to find out: bit.ly/4vc50TK

    #AI #KnowledgeGraphs #LLMs

  42. #RAG gave #LLMs access to external knowledge - but standard retrieval systems still struggle.

    #GraphRAG takes a different approach by using knowledge graphs to map entities, relationships, and provenance. The result?

    Watch the #InfoQ video by Cassie Shum to find out: bit.ly/4vc50TK

    #AI #KnowledgeGraphs #LLMs

  43. gave access to external knowledge - but standard retrieval systems still struggle.

    takes a different approach by using knowledge graphs to map entities, relationships, and provenance. The result?

    Watch the video by Cassie Shum to find out: bit.ly/4vc50TK

  44. Agents do not need RAG or vector databases for most real world work. They need structure and semantics.

    Agent Knowledge Graphs turn mixed repositories of code, docs, configs, and PDFs into a connected model that agents can reason over. This often replaces entire retrieval pipelines.

    antaoalmada.dev/posts/Code-Age

    #AIEngineering #KnowledgeGraphs #CodingAgents #AgentWorkflows #SoftwareArchitecture #Graphify

  45. Agents do not need RAG or vector databases for most real world work. They need structure and semantics.

    Agent Knowledge Graphs turn mixed repositories of code, docs, configs, and PDFs into a connected model that agents can reason over. This often replaces entire retrieval pipelines.

    antaoalmada.dev/posts/Code-Age

    #AIEngineering #KnowledgeGraphs #CodingAgents #AgentWorkflows #SoftwareArchitecture #Graphify

  46. Agents do not need RAG or vector databases for most real world work. They need structure and semantics.

    Agent Knowledge Graphs turn mixed repositories of code, docs, configs, and PDFs into a connected model that agents can reason over. This often replaces entire retrieval pipelines.

    antaoalmada.dev/posts/Code-Age

    #AIEngineering #KnowledgeGraphs #CodingAgents #AgentWorkflows #SoftwareArchitecture #Graphify

  47. 🕒 Can AI keep up with changing facts?
    Knowledge in law, medicine, journalism, and science evolves constantly. As part of #SoftwareCampus, Alexandre Mercier is researching how AI systems can manage time-sensitive information while maintaining consistency and transparency.
    Together with @tu_muenchen and #Datev, the ChronoFact project explores the future of trustworthy knowledge management. 🤖📚⏳
    👉 Learn more: softwarecampus.de/en/projekt/c
    #AI #KnowledgeGraphs #ResearchInnovation #MachineLearning

  48. 🕒 Can AI keep up with changing facts?
    Knowledge in law, medicine, journalism, and science evolves constantly. As part of #SoftwareCampus, Alexandre Mercier is researching how AI systems can manage time-sensitive information while maintaining consistency and transparency.
    Together with @tu_muenchen and #Datev, the ChronoFact project explores the future of trustworthy knowledge management. 🤖📚⏳
    👉 Learn more: softwarecampus.de/en/projekt/c
    #AI #KnowledgeGraphs #ResearchInnovation #MachineLearning

  49. This week, session 08 of #KDAI2026 lecture 08: NLP 04 went live.
    From words to vectors, from vectors to meaning:
    - 🔤 TF-IDF & sparse document vectors
    - 🎲 Naive Bayes classification (spam, sentiment & beyond)
    - 🧠 Neural language models — word2vec, ELMo, BERT
    "You shall know a word by the company it keeps." — J.R. Firth, 1957

    See you, space cowboy… 🤠📡
    #NLP #MachineLearning #KnowledgeGraphs #AI #Word2Vec #BERT @fizise @fiz_karlsruhe @KIT_Karlsruhe

  50. This week, session 08 of #KDAI2026 lecture 08: NLP 04 went live.
    From words to vectors, from vectors to meaning:
    - 🔤 TF-IDF & sparse document vectors
    - 🎲 Naive Bayes classification (spam, sentiment & beyond)
    - 🧠 Neural language models — word2vec, ELMo, BERT
    "You shall know a word by the company it keeps." — J.R. Firth, 1957

    See you, space cowboy… 🤠📡
    #NLP #MachineLearning #KnowledgeGraphs #AI #Word2Vec #BERT @fizise @fiz_karlsruhe @KIT_Karlsruhe