Who is we? Disambiguating the referents of first person plural pronouns in parliamentary debates

TL;DR

This study develops a schema and annotated corpus to disambiguate first person plural pronouns in parliamentary debates, enabling automatic resolution of pronoun referents with data augmentation techniques.

Contribution

Introduces a novel annotation schema and creates an annotated corpus for pronoun disambiguation in political debates, applying weak supervision for data expansion.

Findings

Successful creation of an annotated corpus for pronoun disambiguation

Preliminary results show promise for automatic referent resolution

Data augmentation improves model performance

Abstract

This paper investigates the use of first person plural pronouns as a rhetorical device in political speeches. We present an annotation schema for disambiguating pronoun references and use our schema to create an annotated corpus of debates from the German Bundestag. We then use our corpus to learn to automatically resolve pronoun referents in parliamentary debates. We explore the use of data augmentation with weak supervision to further expand our corpus and report preliminary results.