In a forthcoming paper, we created network models in which we added noise at a specific edge density threshold. For each threshold, we classified the graph models based on their
Betti curves. All analyses and results will be described in a forthcoming paper (expected December 2020).
- Hover over squares to see the number of times the corresponding true model was classified as the corrsponding predicted model.
- Click on a square within the confusion matrix on the left to bring up the corresponding averaged Betti curves on the right.
- Drag the slider on the right to see how changing the threshold parameter affects the classification results.