Towards an Open-source Digital Investigation Platform

Simon Cardoso (1) , Hugo Jean (1) , Martin Cherrier (1) , Adrien Dubettier (1) , Tanguy Gernot (1) , Emmanuel Giguet (1) , Christophe Rosenberger (1)

1 Equipe SAFE - Laboratoire GREYC - UMR6072

Humans generate lots of data in the cyberworlds by their behaviors on social networks, interactions with computers or smartphones (writing documents, capturing images,.. .). Analyzing digital traces from Internet, hard drives or computers is an important issue considering the amount of data to process and the associated applications (criminal investigation, recruitment,.. .). In this paper, we propose an open-source software platform for such a task. It embeds many high-level evaluated tools. These tools rely on recent advances in artificial intelligence and deep learning. Our goal is to facilitate digital investigations conducted by criminal experts, researchers in digital archives, or IT engineers. This platform embeds many tools designed to process a document in order to extract some knowledge (as for example, determining if the filetype has been corrupted). We illustrate its benefit through the analysis of a real memory dump from a hard drive.

Mots clés: Digital investigation; machine learning; software platform; benchmarking; artificial intelligence

Domaines: Recherche d'information [cs.IR]; Intelligence artificielle [cs.AI]; Apprentissage [cs.LG]

Simon Cardoso, Hugo Jean, Martin Cherrier, Adrien Dubettier, Tanguy Gernot, et al.. Towards an Open-source Digital Investigation Platform. 2023 International Conference on Cyberworlds (CW 2023), Oct 2023, Sousse, Tunisia. pp.472-479, ⟨10.1109/CW58918.2023.00078⟩. ⟨hal-04179975⟩