How to decide whther edge is anomalous ?
rohitnitk opened this issue · comments
In the Algorithm, how (on what basis ) you are deciding whether an edge is anomalous or not, given the anomaly score?
(I've read the paper but couldn't find it )
The algorithm currently only outputs anomaly scores and not a binary value whether an edge is anomalous. You can either consider edges with high scores as anomalous or have a threshold based on domain expertise and consider edges having scores higher than that threshold as anomalous.
In the paper, we calculate the ROC-AUC using the anomaly scores generated by the algorithm and the ground truth available.