870,515 Android applications form the base of this dataset, enriched with external malware intelligence. Authors Mohsen and Fadi integrated VirusTotal Premium v3 Enterprise and Quark-Engine 25.3.1 reports, matching 803,320 and 772,732 apps respectively. The enrichment adds 10 new features and revalidates class labels based on VirusTotal detection thresholds.
Use Cases
- Train malware detection classifiers based on the 10 new features from VirusTotal and Quark-Engine.
- Validate existing malware labels using the revalidated datasets based on different VirusTotal detection thresholds.
- Analyze relationships between static application features and external threat intelligence reports.
- Benchmark security analysis tools using the paired VirusTotal and Quark-Engine reports for 772,533 apps.
Strengths
- Large base of 870,515 Android applications with 48 metadata features.
- High match rates: 92.3% of apps linked to VirusTotal reports and 88.8% successfully scanned by Quark-Engine.
- Enrichment adds 10 new parsed features from the two external intelligence sources.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count for the new enriched datasets is unknown, which may limit suitability assessment.
- Last updated 2026-05-18 06:14:34; freshness should be verified.
Provenance
- Source
- Derived from an existing dataset (https://doi.org/10.34894/H0YJFT). Enriched by authors Mohsen and Fadi.
- Collection Method
- Original dataset was enriched using VirusTotal Premium v3 Enterprise and Quark-Engine 25.3.1 intelligence.
- Freshness
- Last updated 2026-05-18 06:14:34