Orbifold Consulting

Illustration: Why property graphs have no reasoning engine Most people don’t notice because they don’t know what it is.

Illustration: rag-graph: graph RAG on plain Postgres It’s a Python library for graphRAG on plain Postgres.

Illustration: Microsoft and Graphs They now have more graph stuff than almost any vendor and yet almost none of it means or feels “graph”.

Illustration: Uncertainty in knowledge graphs Bitemporal annotations and degradation are not enough, there is so much we can’t know for sure, intrinsically or practically.

Illustration: OpenSPG: Ant Group’s semantic-powered graph engine OpenSPG (Semantic-Powered Graph) is an advanced open-source knowledge graph engine developed by the Ant Group (China).

Illustration: Google’s Open Knowledge Format is not a knowledge graph Google just shipped the Open Knowledge Format (OKF): markdown files with YAML frontmatter, cross-linked, version-controlled, git-native.

Illustration: Knowledge graphs and RL The latest being “prompting is dead” or better known as agent loops.

Illustration: TuringDB: git-like versioning for knowledge graphs Most graph versioning conversations stop at temporal tracking: valid-time, transaction-time, bitemporal models that let you ask ‘what did this graph look like last Tuesday’.

Illustration: Graph RAG without pre-built graphs There is an interesting graph RAG paper with the provocative title “You Don’t Need Pre-built Graphs for RAG” TLDR: graph structures should match the reasoning task, not the other way around.

Illustration: SAP and n8n Everyone is talking about the workflow automation angle and the money.

Illustration: Obscure RDF formats Obscure W3C RDF-formats are all around us: RIF, GRDDL, POWDER, SAWSDL, XForms, OWL FULL, CURIE and many more.

Illustration: Knowledge graphs and music A chord progression is a walk on the tonnetz, the toroidal lattice where every triangle is a triad and every edge is a voice-leading move.

Illustration: Ontosphere: a browser-based RDF and OWL editor How can it be that this thing has only 63 stars on Github.

Illustration: Microsoft OmniRAG It’s Azure Cosmos DB extended to support graph traversal as a first-class retrieval strategy alongside vector search and direct database queries.

Illustration: Microsoft Fabric IQ The new Microsoft Fabric IQ semantic layer ships with an ontology of business entities, relationships, rules and actions, a native graph engine for multi-hop reasoning, and an MCP server so AI agents can discover and act…

Illustration: SAP knowledge graph SAP just told the world that “the knowledge graph won.”

Illustration: RDF reification and why RDF-star replaced it Every RDF practitioner has hit this challenge: you have a triple, and you need to say something about that triple (aka metadata).

Illustration: GraphRAG without a knowledge graph When someone says “We beat GraphRAG without a knowledge graph” I’m all ears.

Illustration: Apache TinkerPop: the graph query language nobody talks about Most people learning graph databases start with Cypher or SPARQL.

Illustration: Data follows intent The data should not dictate the ontology, unless the data exactly reflects your focus.

See also this related note on how to use Jaccard in Neo4j via GDS.

import networkx as nx import numpy as np import random import matplotlib.pyplot as plt We will use Erdos-Renyi graphs and the Barabasi-Albert graphs as a basis for our diffusions: N = 1500 k = 20 G1 = nx.erdos_renyi_graph(N, k/N) G2 = nx.barabasi_albert_g…

Prime numbers are the atoms of the numeric world and they can be used to uncover relationships between any pair of numbers.

For many, graph analytics and graph machine learning are synonymous terms.

It’s through simple examples and the basics that one grasps a framework.

The Cora dataset consists of 2708 scientific publications classified into one of seven classes.

Barbell Embedding with PyG The Barbell graph is an interesting graph to look at with respect to node embeddings because it has two distinctive blobs which should get emnedded in latent space in two clusters.

What Is Actually Changing AI is genuinely, measurably displacing routine cognitive work.

⚠️ Work in progress Retrieval-Augmented Generation (RAG) blends retrieval systems with LLMs in order to extend language models beyond their training knowledge.

Marker by Datalab is an open source Python library that can be used to extract structured data from PDFs.

Before transformers were around you needed things like SpaCy to do Named Entity Recognition (NER).

Let’s take a look at the v2 of the NuExtract model we previously visited here.

Nuextract by Numind is dedicated to extraction information in a predefined format.

The following is a classic setup in physics with springs attached to to gliders.

Over the past two years, I’ve explored Retrieval-Augmented Generation (RAG) and its intersection with graph-based systems.

Label propagation is a semi-supervised machine learning algorithm used for classification and community detection tasks on graphs.

Contact us Qwiery is a Graph Application Framework, an open source toolbox to create graph-driven apps.

Keras LSTM Keras has been around for a long time and in the early days I used to assemble networks with it.