Link Prediction Benchmark Results: MRR Scores Across Multiple Datasets

5.5 KB of tabular results for link prediction tasks, measured by Mean Reciprocal Rank (MRR). The data was authored by Yucai Jiang and last updated on June 1, 2026. Results are reported as mean ± standard deviation over five independent runs, with statistical significance indicators.

Use Cases

• Benchmarking new link prediction models against reported MRR scores.
• Analyzing the statistical significance of performance improvements across different datasets.
• Comparing the variance of model performance over multiple runs using the provided standard deviations.

Strengths

• Results include statistical significance indicators (paired t-test, p < 0.05).
• Performance metrics are aggregated over five independent runs, providing a measure of variance.
• Data is shared under a permissive CC-BY-4.0 license.

Limitations

• Row count and specific column names are unknown, limiting detailed assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The 5.5 KB size indicates a very limited scope, likely containing only summary statistics.

Provenance

Source

Yucai Jiang via figshare

Collection Method

Likely contains aggregated experimental results from machine learning evaluations.

Freshness

Last updated 2026-06-01 17:40:56; freshness should be verified.