yuanxiaosc/Entity-Relation-Extraction

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

1230 stars · 268 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A TensorFlow and BERT based pipeline for joint entity and relation extraction, built as a solution for the 2019 Language and Intelligence Competition (SKE) information extraction task. It first classifies relation types with a multi-label classifier, then uses sequence labeling to identify entities, outputting (subject, predicate, object) triples.

Use cases

• extract entity-relation triples from Chinese text
• build a knowledge graph from unstructured sentences
• run schema-constrained information extraction
• fine-tune BERT for relation classification and NER
• reproduce the LIC 2019 SKE competition baseline

When to choose

• you need schema-based Chinese entity and relation extraction with BERT
• you want a reference implementation of a pipeline-style SPO triple extractor
• you are working with the SKE 2019 dataset

When to avoid

• you need a maintained library - the repo has no license and last activity was 2020
• you use PyTorch or modern transformer tooling instead of TensorFlow 1.x
• you need joint (non-pipeline) extraction or non-Chinese languages

Facets

library · maturity maintenance

nlp machine-learning parser machine-learning artificial-intelligence python bert tensorflow entity-relation-extraction information-extraction chinese-nlp sequence-labeling knowledge-graph competition-code natural-language-processing

Member repositories

For agents

markdown · JSON · MCP: product_card(name="yuanxiaosc/Entity-Relation-Extraction")