Recommendation Model Performance Metrics with Accuracy and Cohen's Kappa

Performance metrics for recommendation prediction models, including means and standard deviations over 5 folds. The dataset includes Accuracy (Acc), No Information Rate (NIR), and weighted Cohen's Kappa (CK). It is a 5.5 KB XLS file authored by Christina Kranzinger and last updated on June 4, 2026.

Use Cases

• Benchmarking recommendation algorithms based on reported accuracy and kappa metrics.
• Analyzing model stability based on the standard deviation across 5-fold cross-validation.
• Comparing model performance against a No Information Rate baseline.

Strengths

• Performance metrics are aggregated as a mean over 5 folds, suggesting a cross-validated evaluation.
• Standard deviations are provided, offering insight into metric variability.
• The dataset is openly available under a CC-BY-4.0 license.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

Christina Kranzinger via figshare

Freshness

Last updated 2026-06-04 17:24:39; freshness should be verified.