eRST: A Signaled Graph Theory of Discourse Relations and Organization

TL;DR

eRST introduces an advanced theoretical framework for computational discourse analysis that extends RST to include complex relations, signals, and explainability, supported by tools and a new annotated corpus.

Contribution

The paper presents eRST, a novel extension of RST that handles complex discourse relations and signals, along with tools and a large annotated corpus for analysis.

Findings

Developed a new discourse relation graph framework

Created and evaluated a large annotated corpus

Discussed automatic parsing and applications

Abstract

In this article we present Enhanced Rhetorical Structure Theory (eRST), a new theoretical framework for computational discourse analysis, based on an expansion of Rhetorical Structure Theory (RST). The framework encompasses discourse relation graphs with tree-breaking, non-projective and concurrent relations, as well as implicit and explicit signals which give explainable rationales to our analyses. We survey shortcomings of RST and other existing frameworks, such as Segmented Discourse Representation Theory (SDRT), the Penn Discourse Treebank (PDTB) and Discourse Dependencies, and address these using constructs in the proposed theory. We provide annotation, search and visualization tools for data, and present and evaluate a freely available corpus of English annotated according to our framework, encompassing 12 spoken and written genres with over 200K tokens. Finally, we discuss automatic parsing, evaluation metrics and applications for data in our framework.