You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Currently we are using the Jaccard similarity implementation from Ensmallen - this calculates the Jaccard similarity for two nodes considering only one root node at a time (sort of the 'tree' Jaccard).
In the file we are outputting from semsim, for any pair of nodes, we write out the maximum value observed for the 'tree' Jaccard similarities, considering each root node separately.
What we'd like instead is the Jaccard similarity considering all root nodes at the same time (the 'DAG' Jaccard)
Currently we are using the Jaccard similarity implementation from Ensmallen - this calculates the Jaccard similarity for two nodes considering only one root node at a time (sort of the 'tree' Jaccard).
In the file we are outputting from semsim, for any pair of nodes, we write out the maximum value observed for the 'tree' Jaccard similarities, considering each root node separately.
What we'd like instead is the Jaccard similarity considering all root nodes at the same time (the 'DAG' Jaccard)
@LucaCappelletti94 - how much effort would this be to implement?
@caufieldjh
The text was updated successfully, but these errors were encountered: