Multi-Dataset Validation for Network Intrusion Detection Models

Khorshed Alam published a 5.5 KB Excel file on figshare in April 2026. The dataset supports validation of a novel deep reinforcement learning approach for network intrusion classification. It includes results from testing on benchmark datasets like NF-BoT-IoT, NF-UNSW-NB15, NF-ToN-IoT, NF-ToN-IoT-v2, NF-CSE-CIC-IDS2018, and NF-UNSW-NB15-v3.

Use Cases

• Validating intrusion detection models based on performance across multiple benchmark datasets.
• Testing adaptive modeling strategies based on excluded attack types like DoS and Backdoor.
• Evaluating data balancing techniques like K-means SMOTE for handling class imbalance.
• Comparing deep reinforcement learning approaches to traditional static models for network security.

Strengths

• Multi-dataset validation across six named benchmark network flow datasets.
• Results include evaluation of four specific data balancing techniques: Borderline-SMOTE, SMOTE-ENN, ADYSN, and K-means SMOTE.
• Source code for the underlying methodology is publicly available on GitHub.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The 5.5 KB size suggests the dataset likely contains summary results or metadata, not raw network traffic.

Provenance

Source

Khorshed Alam

Collection Method

Likely contains validation results from experiments on benchmark intrusion detection datasets.

Freshness

Last updated 2026-04-16 17:40:29; freshness should be verified.