Published November 14, 2026 | Version v1.0.0

When 100% Accuracy Is a Warning Sign: analysis code for a condition-blocked audit of ML leak detection

  • 1. Memorial University of Newfoundland
  • 2. Hamad Bin Khalifa University
  • 3. University M'Hamed Bougara of Boumerdes

Description

Analysis and figure-generation code for "Evaluating Near-Perfect Accuracy Claims in Machine-Learning-Based Leak Detection for Two-Phase Pipelines: A Condition-Blocked Audit" (Suleiman, Serpoosh, Elkholy, Ferroudji, Rahman, Hamilton).

The paper is the reference for all reported numbers; this release contains only code. Dependencies are pinned in requirements.txt (Python 3.11.3).

Notes

If you use this code, please cite the paper and this software.

Files

issahnazif-afk/100-percent-warning-sign-v1.0.0.zip

Files (61.3 kB)

Name Size
md5:ac1bc82fbf84b033cc49c1cfd582ec18
61.3 kB Preview Download

Additional details