UNSW-NB15: Cleaned Network Intrusion Detection Data

UNSW-NB15 is a cleaned and optimized dataset for network intrusion detection. The raw description indicates it is a collection of network traffic data, but specific row counts and column details are not provided. The data has been preprocessed, with encoding and scaling left to the user.

Use Cases

• Training binary classifiers to detect malicious network activity based on traffic features.
• Benchmarking anomaly detection algorithms on preprocessed network traffic data.
• Developing feature engineering pipelines for cybersecurity datasets, as encoding and scaling are left to the user.

Strengths

• The dataset has been cleaned and optimized for machine learning tasks, as stated in the description.
• It is a recognized benchmark dataset (UNSW-NB15) for network intrusion detection research.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Last update date is unknown; freshness unverified.

Provenance

Source

Kaggle

Collection Method

Likely derived from the original UNSW-NB15 network traffic capture dataset.