A relational framework for interpreting anaphora
Nand, P; Yeap, W; Nand, P; Yeap, W
MetadataShow full metadata
In this paper we present a novel way to resolve indirect or bridging anaphora which gives us a richer interpretation then the current frameworks. The new framework uses the core set of relations that have been used to describe compound noun generation. We firstly argue that the linguistic processes of compound noun generation and the use of NP anaphora are based on a common relational framework. In order to validate this theory we used human annotators to interpret indirect anaphora from naturally occurring discourses. The annotators were asked to classify the relation between an anaphor and the antecedent into relation types that have been previously used to describe the relation between the modifier and the head noun of a compound noun. We obtained very encouraging results with a Fleiss's kappa value of 0.66 for inter-annotation agreement. This compares well with other similar annotation experiments for relation interpretation in compound nouns. The annotation results strongly indicate that anaphora and noun compounds are based on very similar relational framework, hence should treated the same.