Reproduce “Detecting and Explaining Malware Family Evolution Using Rule-Based Drift Analysis” 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

Olha Jurečková, Martin Jureček

This research from the Czech Technical University in Prague introduces a framework to detect and explain malware family evolution by comparing rule sets generated for original and adversarially modified samples. The method quantifies concept drift using rule distances and identifies specific feature changes, achieving an overall concept drift detection accuracy of 92.08%.

Key insights

  • Rule-based classifiers offer a transparent alternative to black-box models, providing explicit, human-readable explanations for malware characteristics and their changes over time.
  • The magnitude of concept drift can be quantitatively measured by the dissimilarity between rule sets, allowing for objective assessment of evolutionary changes.
  • Comparing feature-level rules derived from malware samples directly reveals which specific attributes and their values have changed, offering granular insights into new evasion tactics.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.