Reproduce “Machine Learning in Digital Forensics: A Systematic Literature Review” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Tahereh Nayerifard, Haleh Amintoosi, Abbas Ghaemi Bafghi, Ali Dehghantanha
A systematic literature review comprehensively analyzed 608 peer-reviewed publications from 2010 to 2021, revealing the exponential growth of machine learning applications in digital forensics. The review found that image forensics accounts for 62.7% of all applications, with Convolutional Neural Networks dominating as the most frequently used technique (38.2%), while also identifying leading research countries and future challenges like explainable AI and adversarial robustness.
Key insights
- Convolutional Neural Networks (CNNs) have become the preferred method for image and video forensics, showing dominance since 2017.
- There is a clear shift in the field from manual feature engineering to automatic feature extraction, primarily using deep learning approaches.
- Image forensics is identified as the most mature application area for machine learning, while domains such as network forensics and data recovery remain relatively underexplored.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.