CodeGraph: Open-Taxonomy Knowledge Graph for Source Code with Wikidata Grounding
Researchers build CodeGraph, a 158-million-node open-taxonomy knowledge graph of source code grounded in Wikidata.
The paper presents a pipeline that uses a code-specialised LLM to annotate source code with an open taxonomy of algorithms, paradigms, design patterns, and application domains, then grounds those concepts in Wikidata. Linking combines deterministic SPARQL, a Deep Research Agent for the long tail, and a hierarchy rollup of parent categories. Run on 167 million Stack-Edu files across 14 languages, CodeGraph contains about 158 million nodes, roughly 63,000 concept entities, about 19,800 Wikidata entities, and approximately 1 billion typed edges. A calibrated QA protocol mixes a small human gold set with an LLM-as-a-judge filter.
- Pipeline extracts algorithms, paradigms, patterns, and domains with a code LLM.
- Wikidata linking uses SPARQL, a research agent, and parent-category rollup.
- Applied to 167 million Stack-Edu files across 14 languages.
- CodeGraph has about 158 million nodes and roughly 1 billion edges.
- Precision is checked with a human gold set and an LLM judge.
Full article236 words · extracted from huggingface.co · click to collapse
Public software repositories, like GitHub and Software Heritage Archive, store billions of files, yet extracting their implicit engineering knowledge ---i.e., the algorithms they implement, the paradigms they follow, the patterns they instantiate, and the application domains they serve--- remains challenging, as current tools are constrained to syntactic and token-level analysis. We present a pipeline for building an open-taxonomy semantic annotation of source code using a code-specialised Large Language Model. The extracted entities are grounded in Wikidata through a three-stage linking procedure: a deterministic SPARQL stage handles unambiguous entities, a Deep Research Agent resolves the residual long tail, and a hierarchy-rollup stage imports the parent-of closure of each resolved Wikidata identifier. The resulting annotations are materialised as a source-code-specific open-taxonomy knowledge graph. We further introduce a calibrated quality-assurance protocol that quantifies annotation precision by combining a small human gold set with an LLM-as-a-judge filter. We applied our pipeline to the 167 million files of the Stack-Edu corpus, creating the first known large-scale open-taxonomy knowledge graph for source code. Our graph, named CodeGraph, contains approximately 158 million nodes, which include around 145 million files, about 63,000 extracted concept entities (such as algorithms, paradigms, design patterns, and application domains), and roughly 19,800 grounded Wikidata entities. Furthermore, CodeGraph features approximately 1 billion typed edges that connect files to their respective concepts, link these concepts to their grounded Wikidata identifiers, and relate them to their parent categories, covering 14 programming languages.
Text extracted automatically; images, tables and formatting may be missing. Original: https://huggingface.co/papers/2609.29474