Per-Class Detection Metrics by APT Phase from 5-Fold Cross-Validation

5.5 KB of per-class detection metrics for Advanced Persistent Threat (APT) phases, calculated using 5-fold cross-validation. The dataset reports mean values with standard deviations and was authored by Adel Alshamrani. It was last updated on June 2, 2026.

Use Cases

• Benchmarking detection models based on reported per-class metrics like precision and recall.
• Analyzing model performance across different APT attack phases mentioned in the description.
• Assessing the stability of detection algorithms based on the provided standard deviation from cross-validation.

Strengths

• Provides aggregated evaluation metrics from a 5-fold cross-validation procedure.
• Includes standard deviation values, offering insight into metric stability.
• Released under a permissive CC-BY-4.0 license for open use.

Limitations

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

Provenance

Source

Adel Alshamrani via figshare

Collection Method

Likely derived from an experiment evaluating a detection model, using 5-fold cross-validation.

Freshness

Last updated 2026-06-02 17:27:32; freshness should be verified.