SEEM SEEM SEEM
  • Home
  • Research
  • People
  • Publications
  • Projects
  • Blog
  • Join / Contact

On this page

  • About
  • Why it matters
  • Key result and audience
  • Public links
  • References

Explainable PhaseNet

Interpreting and strengthening microseismic event detection

Explainable AI
Event detection
Microseismic monitoring
Research context and public links for the Explainable PhaseNet project
Published

June 28, 2026

About

Explainable PhaseNet provides a maintained framework for understanding the waveform evidence behind PhaseNet detections (Abdullin et al. 2026). It implements waveform Grad-CAM, exact component-wise Shapley values, SHAP-gated inference, and controlled noise-robustness experiments.

Why it matters

Event detectors can appear confident for the wrong reasons, especially under pump noise or low signal-to-noise ratios. Explanations help researchers inspect phase sensitivity and use contribution patterns to reduce false alarms.

Key result and audience

The repository reproduces the central manuscript workflows and figures, bundles tested model code, and supplies lightweight synthetic-fixture tests when large datasets are unavailable. It serves microseismic monitoring teams and researchers in trustworthy geoscience AI.

Public links

  • GitHub repository
  • Publication DOI
  • Archived software release
Note

This is an original SEEM-KFUPM repository and is distributed under the MIT License.

References

Abdullin, Ayrat, Denis Anikiev, and Umair Bin Waheed. 2026. “Explainable AI for Microseismic Event Detection.” Artificial Intelligence in Geosciences 7 (3): 100246. https://doi.org/10.1016/j.aiig.2026.100246.

© 2026 Smart Earth Exploration and Monitoring Group

College of Petroleum Engineering & Geosciences

College of Petroleum Engineering & Geosciences · KFUPM

  • Content: CC BY 4.0

  • Powered by Quarto