Paper: A Syntactic and Lexical-Based Discourse Segmenter

ACL ID P09-2020
Title A Syntactic and Lexical-Based Discourse Segmenter
Venue Annual Meeting of the Association of Computational Linguistics
Session Short Paper
Year 2009
Authors

We present a syntactic and lexically based discourse segmenter (SLSeg) that is de- signed to avoid the common problem of over-segmenting text. Segmentation is the first step in a discourse parser, a system that constructs discourse trees from el- ementary discourse units. We compare SLSeg to a probabilistic segmenter, show- ing that a conservative approach increases precision at the expense of recall, while re- taining a high F-score across both formal and informal texts.