Intrusion Detection in IoHT: Network Attack Data for Model Evaluation

A 2.2 GB dataset created by G. Sai Chaitanya Kumar and last updated on 2026-04-21. It contains data for evaluating a lightweight intrusion detection system based on E-GhostNetV2-MobileNeXt, designed to secure the Internet of Health Things (IoHT). The study utilized three widely recognized public datasets for multiclass and binary classification analyses.

Use Cases

• Training machine learning models for binary attack classification based on network traffic data.
• Evaluating multiclass intrusion detection algorithms on recognized public benchmark data.
• Benchmarking lightweight neural network architectures like E-GhostNetV2-MobileNeXt for IoT security.
• Researching attack patterns and reliability in Internet of Health Things (IoHT) environments.

Strengths

• Data volume is 2.2 GB, indicating a medium-scale dataset.
• The model and study are based on three widely recognized public datasets.
• Released under a permissive CC-BY-4.0 license for open use.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

figshare

Collection Method

Likely aggregated or derived from three recognized public datasets for intrusion detection.

Freshness

Last updated 2026-04-21 05:51:22; freshness should be verified.