Please use this identifier to cite or link to this item:
https://hdl.handle.net/2440/66745
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Shi, Q. | en |
dc.contributor.author | Altun, Y. | en |
dc.contributor.author | Smola, A. | en |
dc.contributor.author | Vishwanathan, S. | en |
dc.date.issued | 2007 | en |
dc.identifier.citation | Proceedings of the 2007 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, 2007: pp.640-648 | en |
dc.identifier.uri | http://hdl.handle.net/2440/66745 | - |
dc.description.abstract | In this paper, we study the problem of automatically segmenting written text into paragraphs. This is inherently a sequence labelling problem, however, previous approaches ignore this dependency. We propose a novel approach for automatic paragraph segmentation, namely training Semi-Markov models discriminatively using a Max-Margin method. This method allows us to model the sequential nature of the problem and to incorporate features of a whole paragraph, such as paragraph coherence which cannot be used in previous models. Experimental evaluation on four text corpora shows improvement over the previous state-of-the art method on this task. | en |
dc.description.statementofresponsibility | Qinfeng Shi, Yasemin Altun, Alex Smola and S. V. N. Vishwanathan | en |
dc.language.iso | en | en |
dc.publisher | Association for Computational Linguistics | en |
dc.rights | Copyright 2007 Association for Computational Linguistics | en |
dc.source.uri | http://www.eprints.pascal-network.org/archive/00003986/ | en |
dc.title | Semi-Markov models for sequence segmentation | en |
dc.type | Conference paper | en |
dc.contributor.conference | Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning (2007 : Prague, Czech Republic) | en |
dc.publisher.place | United States | en |
pubs.publication-status | Published | en |
dc.identifier.orcid | Shi, Q. [0000-0002-9126-2107] | en |
Appears in Collections: | Aurora harvest Computer Science publications |
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.