Information Technology Reference
In-Depth Information
the resolution of these types of bridging as well as the traditional co-reference anaphora
at a discourse level. The resulting network of relations between nouns in the discourse
will provide us with an infrastructure which can be utilized in a computational system
for a richer interpretation of discourses.
Fig. 1.
Figure showing the relative distribution of relation types in a corpus of 20 news paper
articles
6
Discussion
The annotation experiment results strongly indicate that the two natural language usage
phenomena of compound noun generation and anaphoric use of nouns are based on the
same underlying semantic structure. At a theoretical level, this has a significant impact
on our understanding of how humans use natural language. In particular, it will help
us better understand the use of compound nouns which are also anaphoric by using the
same theory to interpret them. At a computational level, the proposed framework for
anaphora resolution allows us to marry the two nlp areas so that we can better share
computational advances in the two research areas. Recently there has been an increased
momentum [13,26,20,4,15,21,1,11] towards automatic derivation of relations between
composite nouns in noun compounds, most of them based on relations from [18]. This
will result in an increasing amount of ontology representing semantic relations used for
generating compound nouns. Any such ontology will be directly useful for anaphora
resolution in the framework proposed in this paper.
Another significant advantage of a common framework is that it will be easier to
integrate the full meaning of a compound noun and the meaning associated with it being
used anaphorically. Currently, anaphora is described using a different set of relations
