DOCSEARCHbyAlgolia
Submit Docs

awesome-knowledge-graph

repository·master·Indexed 23 days ago

https://github.com/totogo/awesome-knowledge-graph

A curated collection of resources, tools, and databases related to Knowledge Graphs. Includes infrastructure, graph database implementations (Multi-model, Specialized, Distributed), Triple Stores for RDF data, graph computing frameworks, visualization tools, and knowledge engineering techniques. Provides references for graph query languages (Cypher, Gremlin, SPARQL, GQL), managed hosting services, general and domain-specific datasets, and learning resources for integrating knowledge graphs with LLMs.

Tokens
1.7K
Snippets
0
Records
14
Agent score
34%

What's inside awesome-knowledge-graph

  1. Explore Graph Computing Frameworks

    master

    For large-scale graph analytics and processing, use these frameworks:

    • Distributed Processing: Apache Giraph (iterative processing) and Gradoop (based on Apache Flink).
    • General Purpose/Big Data: Apache Spark - GraphX (graph-parallel computation) and Tencent Plato (distributed graph computation and machine learning).
    • Language-specific Toolkits: RDFLib (Python RDF manipulation), DotNetRDF (.NET RDF library), and Redland C RDF (C libraries).

  2. Explore Graph Construction and Knowledge Engineering tools

    master

    Tools for building and managing knowledge graphs:

    • Construction: Morph-KGC (RML mappings), Ontop (Virtual Knowledge Graph engine for SQL), and Termboard (graphical editor for Terms and Relations).
    • Engineering: KGPrune (extracting subgraphs from Wikidata) and YAGA-NAGA (harvesting/ranking knowledge from the web).
    • Knowledge Fusion (Entity Resolution): Dedupe (Python ML library for fuzzy matching/deduplication) and LIMES (Link Discovery Framework).

  3. Explore Graph Database technologies

    master

    The repository provides a curated list of various graph database implementations categorized by their architecture and use cases. Key categories include:

    • Multi-model Databases: Support for multiple data models (e.g., ArangoDB, ArcadeDB, CosmosDB).
    • Specialized Graph Databases: Including Neo4j (OLTP), Memgraph (In-memory), TigerGraph (Distributed/Parallel), and Kuzu (Embeddable).
    • Type-safe/Language-specific: Atomic-Server (Rust, supports Atomic Data/JSON/RDF) and CogDB (Python).
    • Distributed/Scalable: JanusGraph, Nebula Graph, and HugeGraph (TinkerPop 3 compliant).

  4. Explore Graph Visualization tools

    master

    To visualize graph structures, choose a tool based on your environment:

    • JavaScript/Web: AntV G6 (high performance), Cytoscape.js, Sigma.js (for larger graphs), and KeyLines & ReGraph (React/JS).
    • Python/Data Science: Graphistry (GPU-accelerated, supports Jupyter/Streamlit).
    • Desktop Software: Gephi and Cytoscape (Windows, Mac, Linux).

  5. Explore Triple Stores for RDF data

    master

    Triple stores are specialized databases for storing and querying RDF (Resource Description Framework) data. Options include:

    • High-performance/Enterprise: AllegroGraph (scales to billions of quads), GraphDB (W3C compliant), and Virtuoso.
    • Frameworks: Apache Jena (Java framework for Semantic Web) and Eclipse RDF4J (Java framework for processing RDF data).
    • Lightweight/Specific: Oxigraph (Rust-based) and CLever (SPARQL engine for very large knowledge graphs).

  6. Learn Knowledge Graphs and LLMs

    master
    For developers building modern AI applications, the book 'Knowledge Graphs and LLMs in Action' (Manning Publications) demonstrates how to integrate knowledge graphs constructed from structured and unstructured sources into LLM-powered applications and RAG (Retrieval-Augmented Generation) pipelines.

  7. Learn Graph Computing via Community Resources

    master

    For practical guidance on using graph technologies, refer to these community-driven resources:

    • Graph Book: A TinkerPop3-centric book by Kelvin R. Lawrence.
    • SQL2Gremlin: A guide for transitioning from SQL to Gremlin by Daniel Kuppitz.
    • The Gremlin Compendium: A survival kit for Gremlin users consisting of a 10-part blog series by Doan DuyHai.

  8. Learn Graph Query Languages (Cypher and Gremlin)

    master

    To interact with graph databases, study these official query language documentations:

    • Cypher: Reference documentation for the Cypher query language (commonly used with Neo4j).
    • Gremlin: Reference documentation for the Apache TinkerPop traversal language.

  9. Explore General Knowledge Graph Datasets

    master

    The following datasets provide large-scale, structured, or multilingual knowledge for building and enriching knowledge graphs:

    • BabelNet: A multilingual encyclopedic dictionary and semantic network with ~16 million Babel synsets.
    • Wikidata: A collaborative, multilingual database providing structured data for the Wikimedia movement.
    • Google Knowledge Graph: Millions of entries describing real-world entities (people, places, things).
    • Freebase: A large-scale knowledge base (acquired by Google and used in Google Knowledge Graph).
    • DBpedia: Structured content extracted from Wikimedia projects.
    • XLore: A large-scale English-Chinese bilingual knowledge graph.
    • The GDELT Project: Monitors global news in 100+ languages to identify entities, events, and themes.
    • YAGO: A semantic knowledge base derived from Wikipedia, WordNet, and GeoNames (~10M entities, 120M facts).
    • Zhishi.me: Knowledge Graph data from major Chinese encyclopedias (Baidu Baike, Hudong Baike, Chinese Wikipedia).
    • NELL (Never-Ending Language Learner): Continuously extracts facts from web pages.
    • Golden Protocol: A decentralized, open, and transparent Web 3 knowledge graph.

  10. Reference of Managed Hosting Services for Graphs

    master

    Cloud-managed services for graph databases:

    • AWS: Amazon Neptune and JanusGraph (via DynamoDB backend).
    • Azure: CosmosDB.
    • Google Cloud: JanusGraph (via GKE and Bigtable).
    • Alibaba Cloud: Aliyun Graph Database (beta) and Huawei Cloud Graph Engine Service.
    • Specialized: Neo4j @ Graphene or Graph Story, WoordLift (for SEO/Marmotta), and Baidu Knowledge Graph.

  11. Explore Academic & Research Knowledge Graphs

    master

    These tools provide specialized knowledge graphs for scholarly and scientific information:

    • AMiner: Search and mining services for researcher social networks.
    • Microsoft Academic: Uses machine learning and semantic inference for scholarly exploration.
    • AceMap: Academic search engine for entities like papers, authors, and institutions.
    • Semantic Scholar: AI-powered research tool for scientific literature and scientific knowledge graphs.

  12. Explore Semantic Networks

    master

    Use these resources to understand word meanings and semantic relations:

    • ConceptNet: A free semantic network designed to help computers understand word meanings.
    • Microsoft Concept Graph: Focused on Short Text Understanding.
    • OpenHowNet: An Open Sememe-based Lexical Knowledge Base in Chinese.
    • WordNet: A large lexical database of English from Princeton University.