Comparative Literature Review: Intrusion Detection Datasets for Healthcare IoT Security

Hashim Hussain's study, last updated in May 2026, proposes a Post-Quantum Cognitive Zero-Trust Architecture (PQ-CZTA) for healthcare IoT security. The work is based on an analysis of six intrusion detection datasets: NSL-KDD, CIC-IDS2017, MedBIoT, Edge-IIoTset, IoT-23, and TON_IoT. The 9.5 KB Excel file likely contains comparative results or metadata from this literature review.

Use Cases

• Benchmarking machine learning models for IoT intrusion detection based on the six referenced datasets.
• Analyzing the performance trade-offs of integrating post-quantum cryptography like CRYSTALS-Kyber and SPHINCS+ in low-latency systems.
• Studying the impact of techniques like SMOTE oversampling on classifier performance for imbalanced cybersecurity data.
• Evaluating adaptive, trust-score based policy decisions (ALLOW, MONITOR, DENY, QUARANTINE) for zero-trust architectures.

Strengths

• The study evaluates performance on six distinct, modern intrusion detection datasets, including IoT-specific ones like MedBIoT and Edge-IIoTset.
• Reports specific performance metrics, with F1-scores ranging from 0.972 to 1.000 across the evaluated datasets.
• Includes an ablation study quantifying the contribution of components like SMOTE, which improved F1 scores by 5–20% on imbalanced data.

Limitations

• The 9.5 KB file size suggests the dataset is very small, likely containing summary results or metadata rather than raw intrusion data.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment for large-scale model training.

Provenance

Source

Hashim Hussain via figshare

Collection Method

Likely contains synthesized results from a comparative literature review and analysis of six public datasets.

Freshness

Last updated 2026-05-28 17:42:32; freshness should be verified.