Sign in to view source links and access this dataset
Description
A Chinese-language dataset of supply chain and industry relationship triplets for training relation classification models. It contains approximately 513,000 training samples and 150 test samples, created by author flipped121364 and last updated on Hugging Face in May 2026. Each sample is a triple of [entity1, relation, entity2].
Use Cases
Train relation classification models based on the defined 'upstream', 'downstream', and 'other' relation types.
Fine-tune language models for supply chain knowledge graph construction based on entity-relation pairs.
Benchmark models for Chinese industrial text understanding based on the provided test splits.
Analyze industry interdependencies based on the structured relationship data.
Strengths
Approximately 513,000 training samples provide a substantial base for model training.
Includes two test sets (~120 and 30 samples) for model evaluation.
Relations are clearly defined as 'upstream', 'downstream', and 'other'.
Limitations
Column-level documentation is absent; field semantics must be inferred after download.
Row count for the primary dataset is approximate ('~513,000').
Description metadata is limited; actual data quality requires manual inspection after download.
Provenance
Source
huggingface
Freshness
Last updated 2026-05-11 13:41:35.
Geography
China (implied by Chinese language focus).
License is unknown; terms of use must be verified before application.