MeSH-SNOMED Entity Alignment 15K: A Biomedical Knowledge Graph Benchmark

A benchmark dataset for cross-ontology entity alignment between the Medical Subject Headings (MeSH) and SNOMED CT biomedical terminologies. It contains 15,000 gold-aligned concepts embedded within larger, realistic knowledge graphs containing many non-aligned background entities. The dataset was created by vaibhavalakshmiravideshik and last updated on Hugging Face in May 2026.

Use Cases

• Benchmarking entity alignment algorithms based on the provided gold-standard concept pairs.
• Training models for cross-ontology linking in biomedical informatics based on the MeSH and SNOMED CT graphs.
• Evaluating system performance under realistic large-graph conditions with background entities, as described in the dataset purpose.

Strengths

• Designed for realistic evaluation with gold-aligned concepts embedded in larger graphs containing background entities.
• Provides a specific benchmark scale of 15,000 aligned entities for cross-ontology matching.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and file formats are unknown, which may limit suitability assessment.

Provenance

Source

Hugging Face user vaibhavalakshmiravideshik.

Collection Method

Likely constructed as a benchmark for evaluating entity alignment systems.

Time Range

null

Freshness

Last updated 2026 05 02 01:25:22; freshness should be verified.

Geography

null