The Knowledge Graph Conference Icon
The Knowledge Graph Conference
  • 🏠Home
  • 📅Events
  • 👤Members
  • 🔵Announcements
  • 🔵Ask
  • 🔵Ask The Ontologists
  • 🔵Events
  • 🔵Jobs
  • 🔵Promotions
  • 🔵Share
Powered by Tightknit

Denny V.

Posted in Ask·
Avatar of Denny V.Denny V.
·
·

What is the companion site?

What is the companion site?

See more
0Comments
Commented on New Handbook on Neurosymbolic AI and Knowledge Gra...·Posted inPromotions
Avatar of Denny V.
Denny V.
·

Oh, indeed! I just checked: the English definition focuses on "a reference book that is to be kept handy, at hand, no matter what size", the German definition of "Handbuch", which false-friended me, requires it aim to be "knapp", i.e. concise. My bad! Thanks for making me learn something today

Commented on New Handbook on Neurosymbolic AI and Knowledge Gra...·Posted inPromotions
Avatar of Denny V.
Denny V.
·

Congratulations! With more than 1100 pages, is it really still a handbook? 😉 j/k, well done, and very timely!

1
Commented on Exciting Lead Ontology Engineer Position at LEGO G...·Posted inJobs
Avatar of Denny V.
Denny V.
·

this sounds like a dream job

Commented on Exploring GraphRAG Implementations with RDF Knowle...·Posted inAsk
Avatar of Denny V.
Denny V.
·

A colleague of mine had an interesting approach recently, that combined document and graph RAG-gish ideas (over Wikipedia and Wikidata respectively) for search: https://arxiv.org/abs/2501.11301

1
Posted in Share·
Avatar of Denny V.Denny V.
·

New Research Paper on LLMs, Knowledge Graphs, and Search Engines

Luna Dong (Meta, ex-Amaxon and Google), Aidan Hogan (University of Santiago de Chile), Gerhard Weikum (MPI, Yago) and I published a paper on LLMs, Knowledge Graphs, and Search Engines - https://arxiv.org/abs/2501.06699

See more
0Comments
9
Commented on Publishing Knowledge Graph Datasets in HDT Format:...·Posted inAsk
Avatar of Denny V.
Denny V.
·

One issue we had was that HDT wasn't great in supporting dynamic KGs, which changed often, because it needed a complete recalculation, IIRC. For reasonably static datasets this sounds like a good efficient way.

1

About

  • Job title
  • Location
  • Organization