Toward Fast and Accurate Neural Discourse Segmentation
TL
TL;DR
This paper proposes an end-to-end neural segmenter based on BiLSTM-CRF framework that addresses the problem of data insufficiency by transferring a word representation model that is trained on a large corpus and proposes a restricted self-attention mechanism in order to capture useful information within a neighborhood.
Abstract
Discourse segmentation, which segments texts into Elementary Discourse Units, is a fundamental step in discourse analysis. Previous discourse segmenters rely on complicated hand-crafted features and are not practical in actual use. In this paper, we propose an end-to-end neural segmenter based on BiLSTM-CRF framework. To improve its accuracy, we address the problem of data insufficiency by transferring a word representation model that is trained on a large corpus. We also propose a restricted self-attention mechanism in order to capture useful information within a neighborhood. Experiments on the RST-DT corpus show that our model is significantly faster than previous methods, while achieving new state-of-the-art performance.
References26 items
1
SegBot: A Generic Neural Text Segmentation Model with Pointer Network
2
AllenNLP: A Deep Semantic Natural Language Processing Platform
3
Deep Contextualized Word Representations
4
Gated Self-Matching Networks for Reading Comprehension and Question Answering
5