Entity Relatedness Test Data for Music and Movie Domains

Juana Herrera created this test dataset for evaluating entity relatedness algorithms. It contains 20 entity pairs each from the music and movie domains, with 50 ranked relationship paths per pair. The data is derived from popular reference sources IMDb and last.fm.

Use Cases

• Benchmarking entity relatedness algorithms based on the provided ranked relationship paths.
• Evaluating knowledge graph path ranking methods using the included entity and property relevance scores.
• Testing semantic similarity models for entities in the music and movie domains.

Strengths

• Includes 20 entity pairs from each of two distinct domains (music and movies).
• Provides 50 ranked relationship paths for each entity pair.
• Contains entity ratings and property relevance scores for the paths.

Limitations

• Row count and dataset scale are unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.

Provenance

Source

IMDb and last.fm

Collection Method

Likely extracted and processed from the referenced public datasets.