Harmonized Network Intrusion Detection Data for ML Benchmarking

Shailendra Mishra's framework harmonizes features from the NSL-KDD and CICIDS2017 network intrusion datasets for evaluating machine learning models. The dataset, last updated in April 2026, is a 5.5 KB Excel file containing the harmonized data used in the study. Experimental results from the framework demonstrated a Random Forest model achieving 98.0% accuracy and 97.0% F1-score on this data.

Use Cases

• Benchmarking supervised, unsupervised, and deep learning models for intrusion detection based on harmonized network features.
• Evaluating the impact of feature harmonization on model performance using ablation analysis.
• Studying tamper-evident logging mechanisms for experiment reproducibility in cybersecurity research.

Strengths

• Includes harmonized data from two established network intrusion datasets, NSL-KDD and CICIDS2017.
• Framework results demonstrated high model performance, with 98.0% accuracy and 97.0% F1-score for Random Forest.
• Released under a permissive CC-BY-4.0 license for open use and modification.

Limitations

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

Provenance

Source

Shailendra Mishra via figshare.

Collection Method

Feature harmonization of the NSL-KDD and CICIDS2017 datasets.

Time Range

null

Freshness

Last updated 2026-04 20 17:36:51; freshness should be verified.

Geography

null