NapierOne: Over 500,000 Mixed Files for Cybersecurity Research

Edinburgh Napier University's School of Computing created a modern cybersecurity dataset containing over 500,000 distinct files across 44 popular file types. The dataset was designed to address research reproducibility issues and is intended as a complement to the Govdocs1 dataset. It comprises more than 90 separate data subsets, including examples of file types with high entropy, which is a characteristic of ransomware activity.

Use Cases

• Train and test ransomware detection models based on file entropy characteristics mentioned in the description.
• Benchmark forensic analysis tools using the dataset's over 90 subsets of distinct file types.
• Study file format prevalence and characteristics for cybersecurity research based on the 44 identified common types.
• Improve research reproducibility in cybersecurity by using this standardized, large-scale mixed file collection.

Strengths

• Contains over 500,000 distinct files, providing substantial scale for analysis.
• Covers 44 distinct popular file types, offering broad format representation.
• Structured into more than 90 data subsets for specific file characteristics, such as ZIP compression methods.
• Designed explicitly to address research reproducibility and facilitate replication studies.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Last update date is unknown; freshness unverified.
• The specific source and collection methodology for the 5,000 real-world example files per type is not detailed.

Provenance

Source

School of Computing at Edinburgh Napier University

Collection Method

An alternative consensus approach combined findings from multiple sources on file type usage to create a ranked list, followed by gathering 5,000 real-world examples for each common type.