We have developed a full discourse parser in the Penn Discourse Treebank (PDTB) style. Our trained parser first identifies all discourse and non-discourse relations, locates and labels their arguments, and then classifies their relation types. When appropriate, the attribution spans to these relations are also determined. We present a comprehensive evaluation from both component-wise and error-...

This paper addresses Chinese discourse segmentation based on punctuation mark. Particularly, we propose various kinds of lexical, syntactic, position and punctuation features to train classifiers for Chinese discourse segmentation. Experimental results on CDTB (Chinese Discourse Treebank) show that our method based on punctuation mark is appropriate for Chinese discourse segmentation with 89.2%...

Underspecification-based algorithms for processing partially disambiguated discourse structure must cope with extremely high numbers of readings. Based on previous work on dominance graphs and weighted tree grammars, we provide the first possibility for computing an underspecified discourse description and a best discourse representation efficiently enough to process even the longest discourses...

The goal of this paper is to explore the role of clause subordination in discourse structure. Through the study of the French subordinating conjunction avant que (English before) and its interaction with discourse context, I will attempt to show that clause subordination can affect temporal structure and also discourse structure, by conveying either subordination or coordination between discour...

Discourse markers are words or phrases that speakers use at the beginning of a contribution to signal how it relates to prior discourse. They mark changes in the global discourse structure by e.g. signalling the beginning of a new topic or the return to a previous topic. However, words that are used as discourse markers often also have a sentential function. If discourse markers are to be used ...

Human communication is inherently multimodal. In this study we focus on three channels of spoken discourse: the verbal component, prosody, and gesticulation. We address the question of units that can be identified within these components and in spoken multimodal discourse as a whole. The basic unit of the verbal channel is the clause, reporting an event or a state. A set of prosodic criteria he...

Discourse parsing aims to identify the relationship between different discourse units, where most previous works focus on recovering the constituency structure among discourse units with carefully designed features. In this paper, we propose to exploit Long Short Term Memory (LSTM) to properly represent discourse units, while using as few feature engineering as possible. Our transition based pa...

This study investigated prosodic realization of discourse topic by comparing professional and non-professional speakers’ production of four discourses. In each discourse, the target sentence, as the discourse topic, was either at the discourse initial position or at a non-initial position. Six professional and 20 non-professional speakers read aloud all the discourses with two repetitions. Exte...

This article introduces a qualitative research method called discourse tracing. Discourse tracing draws from contributions made by ethnographers, discourse critics, case study scholars, and process tracers. The approach offers new insights and an attendant language about how we engage in research designed specifically for the critical-interpretive and applied analysis of discourse. More specifi...

The width and the position of the pitch range reveals important information about the structure of a spoken discourse. This paper studies the correlation between the pitch range and the discourse structure based on a large database. The model used to analyze the discourse is based on a two-level description of registers. Primary register features reflect the prosodic phrasing within a discourse...