Syntactic queries and discourse boundaries
Posted: October 3rd, 2013, 3:02 pm
A question for you discourse experts from a query language guy.
I am interested in exploring the extent to which discourse boundaries can be identified using syntactic rules, implementing a clear set of syntactic rules using a query language and a treebank. I'm fine with using a set of rules that is specific to a work (e.g., Mark's discourse boundaries may not be the same as John's).
A few clear examples is sufficient, I'm looking for rules that are likely to have few false positives, or at least a good explanation for an underlying meaning that explains both the use in the discourse boundary and other uses.
Where can I find a good set of candidate rules to play with?
I am interested in exploring the extent to which discourse boundaries can be identified using syntactic rules, implementing a clear set of syntactic rules using a query language and a treebank. I'm fine with using a set of rules that is specific to a work (e.g., Mark's discourse boundaries may not be the same as John's).
A few clear examples is sufficient, I'm looking for rules that are likely to have few false positives, or at least a good explanation for an underlying meaning that explains both the use in the discourse boundary and other uses.
Where can I find a good set of candidate rules to play with?