AppClassNet: A Commercial-Grade Dataset for Application Identification Research

AppClassNet is a commercial-grade dataset designed as a realistic benchmark for traffic classification and management. It was created by Dario Rossi and is accompanied by baseline code for training and testing state-of-the-art models, with details provided in a companion technical report. The dataset is available under an Open Access (green) license.

Use Cases

• Train traffic classification models based on the commercial-grade benchmark data.
• Test application identification algorithms using the provided baseline models.
• Manage network traffic based on realistic application usage patterns described in the dataset.

Strengths

• Described as a commercial-grade dataset, suggesting a high standard of realism.
• Includes companion artifacts with baseline code for model training and testing.
• Has a companion technical report detailing dataset description and baseline model performance.

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

Dario Rossi