Statistical Comparison of Hybrid Architecture vs. DenseNet121 and Swin Transformer V2

Statistical comparison results for a hybrid architecture against DenseNet121 and Swin Transformer V2 based on classification accuracy. Paired tests were conducted across N=10 measurements from two independent runs of 5-fold cross-validation. The dataset was authored by Max Wiedemann and last updated on May 5, 2026.

Use Cases

• Compare classification performance between a hybrid architecture and established models based on statistical significance levels.
• Validate the results of a 5-fold cross-validation procedure using paired statistical tests.
• Assess the statistical significance of accuracy differences in machine learning model comparisons.

Strengths

• Statistical significance levels are explicitly defined with p-values (<0.05, <0.01, <0.001).
• Results are based on N=10 measurements from two independent runs of 5-fold cross-validation.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is very small at 5.5 KB, indicating limited scope.

Provenance

Source

figshare

Collection Method

Statistical analysis of classification accuracy from model comparisons.

Freshness

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