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
Colloquial WordNet uses a hypernym model of adjectives rather than the satellite model of Princeton WordNet. IMHO this is far more sensible and satellites cause all kinds of problems; see #35 for a longer discussion. BUT for this release (2020) we should probably not have 20-odd adjectives that work differently to all the others. This issue will fix these to being similar like in Princeton, we can always go back to hypernyms later
It looks better now. You cannot have a model for 99.9% of the data and another for .1%. For alternative models you have GermaNet and pointers by Sara Mendes.
Colloquial WordNet uses a hypernym model of adjectives rather than the satellite model of Princeton WordNet. IMHO this is far more sensible and satellites cause all kinds of problems; see #35 for a longer discussion. BUT for this release (2020) we should probably not have 20-odd adjectives that work differently to all the others. This issue will fix these to being
similar
like in Princeton, we can always go back to hypernyms laterAffected synsets
The text was updated successfully, but these errors were encountered: