Knowledge Graphs та графи знань

🎓 Key Concepts

RDF

Concept: Resource Description Framework. Subject-Predicate-Object triplets.

Format: XML, Turtle, JSON-LD.

Application: Standard for knowledge graphs.

Ontology

Concept: Formal definition of concepts and relationships.

Standards: OWL, RDFS.

Application: Definition of the structure of a knowledge graph.

Triple Store

Concept: Database for RDF triplets.

Examples: Virtuoso, Blazegraph, GraphDB.

Application: Storage and queries of knowledge graphs.

🔧 Methods of Working with Knowledge Graphs

SPARQL Queries

Concept: Query language for RDF.

Examples: SELECT, CONSTRUCT, ASK queries.

Application: Search and extraction of information.

Knowledge Graph Embeddings

Models: TransE, ComplEx, RotatE, DistMult.

Goal: Representation of entities and relationships as vectors.

Application: Similarity, completion, reasoning.

📚 Practical Examples

Example 1: Creating a Knowledge Graph

Ontology: Define ontology (entities, relationships).

Extraction: Extract entities and relationships from data.

Storage: Store in triple store (GraphDB).

Queries: Use SPARQL for queries.

Example 2: Knowledge Graph Embeddings

Preparation: Prepare triplets (entities, relationships).

Embeddings: Train TransE or ComplEx.

Application: Use for similarity, link prediction.

Try it live

Everything above runs in your browser — open Dimensionality Reduction: PCA, t-SNE & UMAP and change the parameters while it is running. Nothing is installed, nothing is uploaded, the whole model lives in one tab.