A Unified Linear-Time Framework for Sentence-Level Discourse Parsing (2019-05-14T00:00:00.000000Z)

TL;DR

An efficient neural framework for sentence-level discourse analysis in accordance with Rhetorical Structure Theory that comprises a discourse segmenter to identify the elementary discourse units (EDU) in a text, and a discourse parser that constructs a discourse tree in a top-down fashion.

Abstract

We propose an efficient neural framework for sentence-level discourse analysis in accordance with Rhetorical Structure Theory (RST). Our framework comprises a discourse segmenter to identify the elementary discourse units (EDU) in a text, and a discourse parser that constructs a discourse tree in a top-down fashion. Both the segmenter and the parser are based on Pointer Networks and operate in linear time. Our segmenter yields an F1 score of 95.4%, and our parser achieves an F1 score of 81.7% on the aggregated labeled (relation) metric, surpassing previous approaches by a good margin and approaching human agreement on both tasks (98.3 and 83.0 F1).

Authors

Shafiq R. Joty

16 papers

Xiang Lin

2 papers

M Saiful Bari

6 papers

TL;DR

Abstract

Authors

References35 items

Toward Fast and Accurate Neural Discourse Segmentation

SegBot: A Generic Neural Text Segmentation Model with Pointer Network

Stack-Pointer Networks for Dependency Parsing

Deep Contextualized Word Representations

A Two-Stage Parsing Method for Text-Level Discourse Analysis

Attention is All you Need

Get To The Point: Summarization with Pointer-Generator Networks

Cross-lingual RST Discourse Parsing

Deep Biaffine Attention for Neural Dependency Parsing

Discourse Parsing with Attention-based Hierarchical Neural Networks

CODRA: A Novel Discriminative Framework for Rhetorical Analysis

Pointer Networks

GloVe: Global Vectors for Word Representation

Recursive Deep Models for Discourse Parsing

Sequence to Sequence Learning with Neural Networks

Neural Machine Translation by Jointly Learning to Align and Translate

Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation

Representation Learning for Text-level Discourse Parsing

A Linear-Time Bottom-Up Discourse Parser with Constraints and Post-Editing

A Novel Discriminative Framework for Sentence-Level Discourse Analysis

Text-level Discourse Parsing with Rich Linguistic Features

Varieties of Question Generation: Introduction to this Special Issue

Analysis of Discourse Structure with Syntactic Dependencies and Data-Driven Shift-Reduce Parsing

The utility of parse-derived features for automatic discourse segmentation

Discourse Chunking and its Application to Sentence Compression

Sentence Level Discourse Parsing using Syntactic and Lexical Information

The rhetorical parsing of unrestricted texts: a surface-based approach

The automatic construction of large-scale corpora for summarization research

Long Short-Term Memory

Building a Large Annotated Corpus of English: The Penn Treebank

Rhetorical Structure Theory: Toward a functional theory of text organization

Using Discourse Structure Improves Machine Translation Evaluation

HILDA: A Discourse Parser Using Support Vector Machine Classification

RST Discourse Treebank (RST– DT) LDC2002T07

editors

Field of Study

Venue Information

Name

Type

URL

Alternate Names