Overview
Reviewing DORA’s visual output graphs is key to understanding model performance, spotting errors, and seeing which features influence predictions. This helps you refine results and align them with your exploration goals.
This article includes interactive walkthroughs for each output graph. For more detail on interpreting and adjusting results, follow the linked step-by-step articles.
ROC Curve (legacy output; only available in older experiments that have not been re-run)
VRIFY Prospectivity Score (VPS)
The VPS is a core output of a DORA Prediction Map. It represents the AI model’s calculated probability of finding your desired commodity, at your specified grades, within the Area of Interest (AOI). The VPS helps geoscientists visually interpret mineral potential across a project site, supporting data-driven exploration decisions.
🔗 Read the full article: What is the VRIFY Prospectivity Score?
Prediction Accuracy (Confusion Matrix)
These outputs help you evaluate how well the model distinguishes between mineralized and barren areas, and what to do if results are not ideal.
🔗 Read the full article: Interpret DORA Results: Prediction Accuracy (Confusion Matrix).
Depth Accuracy (R² Valid)
These outputs help you evaluate how well the model predicts the vertical position of mineralized targets, and what to do if results are not ideal.
🔗 Read the full article: Interpret DORA Results: Depth Accuracy (R² Valid).
Feature Importance (SHAP Values)
This graph helps you understand which features (such as geology, geophysics, or geochemistry) influenced the model’s predictions, and what to do if the model appears to be focusing on the wrong inputs.
🔗 Read the full article: Interpreting DORA Results: Feature Importance (SHAP).
ROC Curve
Note: The ROC Curve is no longer included in new DORA outputs. This section is provided for users reviewing legacy experiments.
The ROC (Receiver Operating Characteristic) curve measures how well the model separates mineralized from barren zones. It plots the True Positive Rate (Sensitivity) against the False Positive Rate (1 - Specificity) across different classification thresholds.
A curve close to the top-left corner indicates strong performance, or high detection of mineralized zones with few false positives.
A curve near the diagonal line suggests poor performance and demonstrates that the model isn’t effectively distinguishing between classes.
The AUC (Area Under the Curve) summarizes this performance:
1.0 = perfect prediction
0.5 = no better than random
Learn More
Create a Prediction Map:
Still Have Questions?
Reach out to your dedicated DORA contact or email support@VRIFY.com for more information.