Cross-dataset Generalization: Harmonized Network Intrusion Detection Data

A 5.5 KB Excel dataset created by Shailendra Mishra and last updated on April 20, 2026. It contains harmonized features from the NSL-KDD and CICIDS2017 network intrusion datasets, processed through a unified framework for evaluating machine learning-based Intrusion Detection Systems (IDS). The work includes results from benchmarking supervised, unsupervised, deep learning, and ensemble models.

Use Cases

• Benchmarking machine learning models for intrusion detection based on harmonized network traffic features.
• Evaluating the impact of feature harmonization on model performance across different datasets.
• Conducting statistical validation of IDS models using tests like Wilcoxon signed-rank and McNemar's.
• Studying cross-dataset generalization in cybersecurity using legacy and contemporary network data.

Strengths

• Dataset is licensed under CC-BY-4.0, permitting open sharing and adaptation.
• The described framework achieved 98.0% accuracy and 97.0% F1-score with a Random Forest model on the harmonized data.
• Includes a proposed cryptographic logging mechanism using SHA-256 hash chaining for tamper-evident traceability.

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 is based on manual alignment and may not be effective for highly heterogeneous datasets.

Provenance

Source

figshare

Collection Method

Features were harmonized from the NSL-KDD and CICIDS2017 datasets via a preprocessing pipeline.

Time Range

The temporal coverage of the original source datasets (NSL-KDD and CICIDS2017) is not specified.

Freshness

Last updated 2026-04-20 17:36:45; freshness should be verified.

Geography

Spatial coverage is not specified.