Interpreting Anaphoric Shell Nouns using Antecedents of Cataphoric Shell Nouns as Training Data

This work proposes an approach that uses automatically extracted antecedents of CSNs as training data to interpret ASNs, and achieves precisions in the range of 0.35 to 0.72, depending upon the shell noun.

Abstract

Interpreting anaphoric shell nouns (ASNs) such as this issue and this fact is essential to understanding virtually any substantial natural language text. One obstacle in developing methods for automatically interpreting ASNs is the lack of annotated data. We tackle this challenge by exploiting cataphoric shell nouns (CSNs) whose construction makes them particularly easy to interpret (e.g., the fact that X). We propose an ap-proach that uses automatically extracted antecedents of CSNs as training data to interpret ASNs. We achieve precisions in the range of 0.35 (baseline = 0.21) to 0.72 (baseline = 0.44), depending upon the shell noun. 1