Hi,
I am trying out some simple rule based feature extraction using Ruta. In one of
the cases I want to mark up places where certain phrases occur in the beginning
of a sentence, and I want to keep the phrases in a list that I load as resource
and then apply using the MARKFAST action, but I don’t know how to constrain it
to sentence beginnings only. Doing something like
Sentence { STARTSWITH(Sentence) ->
MARKFAST(MyFeatureAnnotation, phraseList)
}
will obviously not work because the sentence always starts with a sentence, and
it doesn’t seem to put any constrain on the MARKFAST action itself. I can then
try to unmark those again that don’t align with a sentence start, but it’s
kinda clumsy, and I have to be careful not to unmark other MyFeatureAnnotation
created previously by other rules/annotators. It’s doable but I like to
understand if there is a smarter way, which has so far eluded me after looking
at the documentation?
Any input is appreciated, thanks :)
Cheers,
Mario Juric
