Does syntax help discourse segmentation? Not so much
TL
TL;DR
The results show that dependency information is less useful than expected, but the model provided is a fully scalable, robust model that only relies on part-of-speech information, and it performs well across languages in the absence of any gold-standard annotation.
Abstract
Discourse segmentation is the first step in building discourse parsers. Most work on discourse segmentation does not scale to real-world discourse parsing across languages, for two reasons: (i) models rely on constituent trees, and (ii) experiments have relied on gold standard identification of sentence and token boundaries. We therefore investigate to what extent constituents can be replaced with universal dependencies, or left out completely, as well as how state-of-the-art segmenters fare in the absence of sentence boundaries. Our results show that dependency information is less useful than expected, but we provide a fully scalable, robust model that only relies on part-of-speech information, and show that it performs well across languages in the absence of any gold-standard annotation.
References42 items
1
Cross-lingual and cross-domain discourse segmentation of entire documents
2
DyNet: The Dynamic Neural Network Toolkit
3
Transition-Based Dependency Parsing Exploiting Supertags
4
UDPipe: Trainable Pipeline for Processing CoNLL-U Files Performing Tokenization, Morphological Analysis, POS Tagging and Parsing
5