Representing Discourse Coherence: A Corpus-Based Study

Abstract

This article aims to present a set of discourse structure relations that are easy to code and to develop criteria for an appropriate data structure for representing these relations. Discourse structure here refers to informational relations that hold between sentences in a discourse. The set of discourse relations introduced here is based on Hobbs (1985). We present a method for annotating discourse coherence structures that we used to manually annotate a database of 135 texts from the Wall Street Journal and the AP Newswire. Alltexts were independently annotated by two annotators. Kappa values of greater than 0.8 indicated good interannotator agreement. We furthermore present evidence that trees are not a descriptively adequate data structure for representing discourse structure: In coherence structures of naturally occurring texts, we found many different kinds of crossed dependencies, as well as many nodes with multiple parents. The claims are supported by statistical results from our hand-annotated database of 135 texts.

A Study of Optimized Summarization Models on Mobile Devices for Sight-Interpreters

“…The first way to extract English summary can be called sentence-based summarization which includes a multi-documental summarization system developed successfully by Rafael Ferreira et al. [2] The general idea of a large text is automatically extracted through sentence clustering extraction technique by linguistic treatment on extracting generalization structure categorized by discourse relations studied by Wolf & Gibson [3] . A centroid-based summarization of topic sentence clusterings was developed by Dragomir R. Radev et al, [4] through detecting and retrieving topic of large documents for summarization and archetypal sentences summarization proposed and studied by Ercan Canhasi [5] to produce effective summaries.…”

Discourse oriented summarization

“…The hierarchical nature of coherence was recently challenged by Wolf and Gibson (2005). They argue that the presence of crossed dependencies and nodes with multiple parents render the tree representation of discourse structure inappropriate.…”

“…multiple compatible trees representing 'parallel' interpretations. Ambiguity and simultaneous analyses are not discussed in Wolf and Gibson (2005).…”

“…Other intention-based discourse models are those of Grosz and Sidner (1986), Wolf and Gibson (2005). Their position is that a text is intended to convey a piece of information or an opinion, and that the author uses most of the text to support that information.…”

“…In their evaluation measurements, the dynamic decision-based model achieved near-human performance. The representation of discourse as a hierarchical tree is disputed by Wolf and Gibson (2005). They built a summarization system which used their non-hierarchical discourse model for content selection (Wolf and Gibson, 2004).…”