Binary Classification Performance for Logistic Regression and Bayesian Models

A 5.5 KB dataset by Lukas Waltenberger, last updated on 2026-05-05, contains performance measures for binary logistic regression and Bayesian models. The analysis employed post-hoc filtering, excluding subjects with a sex prediction probability below 75% to refine model evaluation. The dataset is shared under a CC-BY-4.0 license on figshare.

Use Cases

• Comparing classification performance metrics between binary logistic regression and Bayesian models based on the described evaluation.
• Studying the impact of post-hoc filtering on model performance metrics based on a 75% probability threshold.
• Benchmarking model evaluation techniques for binary classification tasks as described in the dataset.

Strengths

• Dataset is small (5.5 KB), facilitating quick download and inspection.
• The description specifies a concrete methodological detail: subjects with a sex prediction probability below 75% were excluded from analysis.
• The dataset has a clear license (CC-BY-4.0) and author attribution (Lukas Waltenberger).

Limitations

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

Provenance

Source

figshare

Freshness

Last updated 2026-05-05 17:31:04; freshness should be verified.