Traditional RAG can't answer "show me all contracts affected by clause 7.3 changes."
GraphRAG can.
Relationships matter.
Let's talk!
#GraphRAG #KnowledgeGraph #EnterpriseAI #dougortiz #postgres
#knowledgeGraph
Today, at the NFDI-MatWerk AHoD Meeting in Siegburg: @enorouzi is giving a demo on out MSE-KG, the knowledge graph-based infrastructure for research data of the consortium.
The MatWerk KG: https://nfdi.fiz-karlsruhe.de/matwerk/
#ontologies #knowledgegraph @NFDI #NFDIMatWerk #researchdata #NFDIrocks #Siegburg #materialsscience @fiz_karlsruhe @nfdi4culture @NFDI4DS
New publication:
Recursive Identity: Structural Conditions of Emergent Continuity – A Theoretical Monograph (Version 1.2)
A curated, revised, and extended international version of the original German Master of Record is now available.
https://doi.org/10.5281/zenodo.18924601
#RecursiveIdentity #Identity #AI #ArtificialIntelligence #SystemTheory #Ontology #HybridArchitecture #ComputationalPhilosophy #MetaResearch #KnowledgeStructures #KnowledgeGraph #DigitalScholarship #OpenScience #Zenodo #DOI
📰 GraphRAGを実際に構築して分かった「使うほど賢くなるAI」の仕組み (👍 30)
🇬🇧 Explains how GraphRAG makes AI smarter by connecting knowledge through graphs, beyond simple document retrieval in traditional RAG systems
🇰🇷 전통적 RAG의 한계를 넘어 GraphRAG가 지식 그래프로 정보를 연결해 AI를 더 똑똑하게 만드는 원리 설명
Ilir Aliu (@IlirAliu_)
Obsidian 노트의 임베딩을 활용해 '사고의 형태'를 보여주는 3D 사고지도를 만들었다는 소개입니다. Obsidian에서 추출한 노트 임베딩을 3D 네트워크로 배치해 중앙화·분산·분포된 사고 구조를 탐색할 수 있도록 시각화하는 응용 사례를 설명합니다.
⭐ DEADLINE EXTENSION ⭐
If you are working on exciting ideas, innovative systems, or early-stage research that deserves visibility, this is your chance to showcase your work and engage with the community.
We have extended the deadline for poster and demo papers to March 10.
🚀 CfP: https://lnkd.in/dtyVUWjF
#SemanticWeb #ESWC2026 #PostersandDemos #CallforPapers #KnowledgeGraph #AI #LLM
A new snapshot release of the #dblp data set has been published. #opendata #linkeddata #knowledgegraph
https://drops.dagstuhl.de/entities/collection/dblp
https://doi.org/10.4230/dblp.rdf.ntriples
https://doi.org/10.4230/dblp.xml
The promise of #agenticAI is being realized in systems like the Service Copilot that Zeiss microscopes provides for its field service engineers, integrating technical documentation, subject matter expertise, and user-generated insights, all orchestrated and shared with a suite of AI agents.
While it relies heavily on modern LLM technology, Max Gärber says it's the system's solid #knowledgeGraph and metadata foundation that make it a success.
Wer mal richtig mit Musik 🎶 , KI 🤖 und dem Culture Knowledge Graph 🐙 hacken will, ist ganz herzlich zur Teilnahme an der Data Challenge von @hermes_datenkompetenzzentrum und @nfdi4culture eingeladen! Das Ziel lautet "Develop an AI-driven tool, allowing researchers to explore musicological knowledge in the Culture Knowledge Graph and beyond conversationally".
📅 Das erste Kick-Off-Event findet am 18.03.26 statt
➡️ Weitere Infos: https://hermes-hub.de/forschen/datachallenges/challenges/challenge-2026.html
ZOYA ✪ (@HeyZoyaKhan)
Obsidian 노트를 임베딩으로 변환해 3D 사고 지도(3D thought map)로 시각화한 사례입니다. 임베딩을 '라이브 네트워크'로 표현해 중앙집중형(핵심 아이디어), 분산형(주제별 클러스터), 분포형(AI 라벨링된 연결) 등 패턴으로 아이디어 연결을 시각적으로 확인할 수 있음을 보여줍니다.
https://x.com/HeyZoyaKhan/status/2024553026451886330
#embeddings #visualization #obsidian #mindmapping #knowledgegraph
FYI: Server Analysis: DevOps, Cooking, Travel & Sci-Fi #shorts: Discover how a semantic knowledge graph connects the term 'server' to DevOps, cooking, travel, and sci-fi by analyzing data and identifying semantically similar terms. It's all about data! #semantic #knowledgegraph #DevOps #cooking #travel #scifi https://www.youtube.com/shorts/8g-mZm78kss
Abhishek Yadav (@abhishek__AI)
Rowboat는 이메일과 회의를 로컬에서 살아있는 지식 그래프로 변환하는 오픈소스 AI 툴입니다. 사용자가 모델을 제공(BYO, Ollama 또는 API)하면 백그라운드 에이전트가 일일 실행되어 과거 결정 기반의 회의 요약을 만들고, 문맥을 활용해 프레젠테이션 데크와 PDF 슬라이드를 자동으로 생성하는 워크플로우를 제공합니다.
New theoretical release now available via Zenodo:
https://doi.org/10.5281/zenodo.18583941
#AI #ArtificialIntelligence #SystemTheory #HybridArchitecture
#OpenScience #Ontology #Zenodo #OSF #Preprint #DOI
#KnowledgeStructures #MetaResearch #ComputationalPhilosophy
#KnowledgeGraph #AIPhilosophy #DigitalScholarship
ERP нового поколения. Объединение мультиагентных AI-платформ и микросервисной архитектуры
Архитектуры корпоративных ИТ-систем стоят на пороге структурных изменений. Их причина - в совмещении двух подходов, которые долгое время развивались параллельно - микросервисные архитектуры и AI-агентные платформы. Микросервисы стали стандартом построения масштабируемых и надежных корпоративных систем, особенно для таких классов, как ERP-системы, для которых важны детерминизм, воспроизводимость результатов, контроль и соответствие регуляторным требованиям [1][2][5]. AI-агентные платформы, в свою очередь, дают возможность системам работать с целями, автономно рассуждать и принимать решения в условиях неопределенности. Попытки использовать эти подходы по отдельности приводят либо к жестким и плохо адаптирующимся системам, либо к интеллектуальным решениям, которым бизнес не готов доверять. В этой статье впервые представлена архитектурную модель, которая позволяет объединить эти два подхода в единую управляемую архитектуру.
https://habr.com/ru/companies/mt-integration/articles/994898/
#ERP #АрхитектураПО #Микросервисы #AI #AIагенты #EnterpriseArchitecture #SystemDesign #DistributedSystems #DigitalTransformation #KnowledgeGraph
💙 Love Data Week 2026 – Dienstag
📢 Where´s the Data?” - Im Knowledge Graphen von #NFDI4Objects natürlich!
Auch heute gibt es zur diesjährigen Love Data Week unter dem Motto: “Where´s the Data?” etwas zum Finden.
✨ Um die Frage des Mottos zu beantworten, möchten wir Euch gerne unseren Knowledge Graphen vorstellen:
🔗 N4O Knowledge Graph: https://www.nfdi4objects.net/portal/services/nfdi4objects-graph/
#N4OGraph #KnowledgeGraph #LinkedOpenData #DH #NFDIrocks #DigitalesKulturerbe #Objektbiografie #LoveDataWeek #WheresTheData
RADAR ist mit seinem Wissensgraphen seit Kurzem im Knowledge Graph Registry des NFDI-Basisdienstes #KGI4NFDI (Knowledge Graph Infrastructure for NFDI) vertreten.
Damit wird der RADAR-Wissensgraph in die gemeinsame semantische Infrastruktur der #NFDI eingebunden und in deren Kontext sichtbar gemacht.
Details unter https://radar.products.fiz-karlsruhe.de/de/nachricht/radar-und-kgi4nfdi-einbindung-des-radar-wissensgraph-den-nfdi-basisdienst
#Forschungsdaten #FAIRdata #Wissensgraph #KnowledgeGraph #FDM #openscience
A new snapshot release of the #dblp data set has been published. #opendata #linkeddata #knowledgegraph
https://drops.dagstuhl.de/entities/collection/dblp
https://doi.org/10.4230/dblp.rdf.ntriples
https://doi.org/10.4230/dblp.xml
Cần hướng dẫn tạo **kết cấu tri thức (KG)** từ hàng trăm tài liệu PDF về quy định vận hành máy móc và nghiên cứu trường hợp. Mục tiêu: liên kết các sự kiện (quy trình lắp đặt, thông số), sử dụng **LLM** để tổng hợp kiến thức và trả lời câu hỏi mở. Gợi ý: dùng công cụ như Neo4j, OpenRefine, hoặc GraphRAG để trích xuất thực thể và mối quan hệ từ PDF không cấu trúc. #KnowledgeGraph #LLM #DataEngineering #AI #TriThuc #KienTrucDuLieu #AIVietNam
(NOTE: Post is in Vietnamese with bilingual tags, und
#KnowledgeGraph #metadata
New #dataset: GESIS Knowledge Graph Version 2.0.0 released
The GESIS Knowledge Graph (GESIS KG) represents metadata of scientific resources available in the GESIS Search and its semantic relationships in an integrated and consistent form. By using established standards, GESIS KG ensures interoperability and reusability of GESIS metadata and links on the Web.
Công cụ mã nguồn mở **Eleutherios** hỗ trợ phân tích tri thức qua đồ thị kết nối, phát hiện xung đột và kiểm duyệt thông tin bằng cách trích xuất luận điểm nguyên tử. Sử dụng Neo4j, PostgreSQL và Ollama chạy cục bộ, không phụ thuộc đám mây. Ưu điểm: phát hiện dấu hiệu ức chế thông tin, mẫu hợp tác đa nguồn, mâu thuẫn/đồng thuận qua dữ liệu. Hiện đã có MVP, xử lý 47K luận điểm, MIT License. #CôngNghệ #MãNguồnMở #TríTuệNhânTạo #OpenSource #AI #KnowledgeGraph
Client Info
Server: https://mastodon.social
Version: 2025.07
Repository: https://github.com/cyevgeniy/lmst