The Popular Content Filenames Dataset: Deriving Most Likely Filenames from the Software Heritage Archive - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2023

The Popular Content Filenames Dataset: Deriving Most Likely Filenames from the Software Heritage Archive

Résumé

The Popular Content Filenames Dataset provides for each unique file content present in the Software Heritage Graph dataset its most popular filename. For the 2022-04-25 version, it contains over 12 billion entries and weights 413 gigabytes. This dataset allows to easily select subsets of the file contents from the Software Heritage archive based on file name patterns, facilitating reseach tasks in areas like data compression and machine learning.
Fichier principal
Vignette du fichier
main.pdf (185.89 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04171177 , version 1 (26-07-2023)

Licence

Paternité

Identifiants

  • HAL Id : hal-04171177 , version 1

Citer

Valentin Lorentz, Roberto Di Cosmo, Stefano Zacchiroli. The Popular Content Filenames Dataset: Deriving Most Likely Filenames from the Software Heritage Archive. 2023. ⟨hal-04171177⟩
107 Consultations
26 Téléchargements

Partager

Gmail Facebook X LinkedIn More