BrambleXu/knowledge-graph-learning

What this project is

Knowledge Graph Learning Resources is a curated link collection — an "awesome list" — for anyone studying or building knowledge graphs (a way of organizing information as connected facts, e.g. "Albert Einstein" → "was born in" → "Ulm"). The project gathers papers, tutorials, datasets, code tools, and community links so learners don't have to hunt for them across the web.

What you'll find inside

The README is organized into several sections:

Section	What it contains
Papers and Notes	Research papers and reading notes, organized by task (e.g., named entity recognition, relation extraction, question answering) and by model type (e.g., BERT, graph neural networks). Notes are tagged with GitHub issue labels; most are in Chinese, but other languages are welcome.
Tutorials, Surveys, and Slides	Introductory guides, academic survey papers, and slide decks — many from major conferences (KDD, WSDM) and companies (Neo4j, IBM, Yahoo). Several entries are in Chinese.
Courses and Lectures	A Chinese-language course for building knowledge graphs from scratch (via a Zhihu column).
Datasets	Public datasets for training and testing knowledge graph systems — e.g., TACRED, SemEval-2010, biomedical knowledge graphs, and GraphRAG benchmarks.
Projects and Implementations	Working example systems, such as a movie knowledge graph with question answering, and newer LLM-based frameworks like LightRAG and HippoRAG.
Libraries and Tools	Software building blocks: graph databases (Jena, Oxigraph), Python libraries (RDFLib, PyKEEN, OpenKE), information-extraction toolkits (DeepKE, OpenNRE), and annotation tools (doccano).
Communities	Places to connect with other practitioners, including China's OpenKG community.

Key details worth knowing

• Bilingual focus: Resources are in both English and Chinese. When a title doesn't make the language obvious, the README marks it explicitly.
• How notes are organized: Paper notes live inside the project's GitHub issues, tagged with labels (e.g., NER(T) for named entity recognition, GNN(M) for graph neural networks). Clicking a link filters the issues to that topic.
• Dates mean "last update," not "publication": For each entry, the year shown is the most recent meaningful update — for maintained projects, that's recent repository activity; for static papers, that's the publication or revision year. Entries are sorted newest to oldest.
• Links were verified on August 26, 2026.
• Contributions are welcome, but strictly regulated: no commercial promotion, no affiliate links, no self-promotion, and each entry must be stable, relevant, and publicly accessible.

What this project is not

• It is not a course or a tutorial itself — it's a directory pointing to other resources.
• It is not software — there's no code to run or install.

Bottom line

If you want to learn about knowledge graphs — from the underlying NLP tasks to graph databases and modern LLM-based systems — this list gives you a high-quality, maintained starting point, especially if you read Chinese or want access to Chinese-language materials.