Evaluation Metrics Summary for Intrusion Detection System Benchmarking

Shailendra Mishra's evaluation metrics reporting summary, published on figshare in April 2026. The 5.5 KB XLS file contains results from a unified framework for evaluating Intrusion Detection Systems (IDS). The framework harmonized features from the NSL-KDD and CICIDS2017 datasets and benchmarked supervised, unsupervised, deep learning, and ensemble models.

Use Cases

• Benchmarking machine learning models for intrusion detection based on reported accuracy and F1-scores.
• Analyzing the impact of feature harmonization on model performance using the described ablation analysis.
• Comparing statistical validation methods for IDS evaluations based on the Wilcoxon, McNemar, and DeLong tests mentioned.
• Studying reproducibility frameworks for cybersecurity experiments based on the described cryptographic logging mechanism.

Strengths

• Provides specific performance metrics, including 98.0% accuracy and 97.0% F1-score for a Random Forest model.
• Framework is explicitly designed for reproducibility and statistical validation, addressing noted challenges in the field.
• Results are derived from harmonizing two established network intrusion datasets, NSL-KDD and CICIDS2017.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The description notes the feature harmonization approach is manual and might not be effective for highly heterogeneous datasets.

Provenance

Source

Shailendra Mishra

Collection Method

Results from an experimental framework integrating feature harmonization, multi-model benchmarking, and statistical validation.

Time Range

null

Freshness

Last updated 2026-04-20 17:36:54; freshness should be verified.

Geography

null