Skip to main navigation Skip to search Skip to main content

Piecewise large margin learning for partially annotated sequences

  • City University of Hong Kong
  • Harbin Institute of Technology Shenzhen

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Supervised and semi-supervised sequence labeling methods require large amounts of fully annotated training sequences or exact annotations of structured outputs. The problem of learning from partially annotated sequences arises in many applications, for example, Natural Language Processing and Computational Biology. In this paper, we propose Piecewise Convex Learning from Partial Labels (PW-CLPL) which is an effective discriminative structured learning method for sequence labeling as global training is intractable for partially annotated sequences. A small number of constraints is reformulated for the optimization, which improve the efficiency of parameters learning. Experimental results on the reconstructed dataset CoNLL-2000 show the effectiveness of the proposed model in the setting of partial annotations.

Original languageEnglish
Title of host publicationProceedings - 2019 IEEE 17th International Conference on Industrial Informatics, INDIN 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1483-1487
Number of pages5
ISBN (Electronic)9781728129273
DOIs
StatePublished - Jul 2019
Externally publishedYes
Event17th IEEE International Conference on Industrial Informatics, INDIN 2019 - Helsinki-Espoo, Finland
Duration: 22 Jul 201925 Jul 2019

Publication series

NameIEEE International Conference on Industrial Informatics (INDIN)
Volume2019-July
ISSN (Print)1935-4576

Conference

Conference17th IEEE International Conference on Industrial Informatics, INDIN 2019
Country/TerritoryFinland
CityHelsinki-Espoo
Period22/07/1925/07/19

Keywords

  • Partial annotation
  • Partial label learning
  • Piecewise learning
  • Sequence labeling

Fingerprint

Dive into the research topics of 'Piecewise large margin learning for partially annotated sequences'. Together they form a unique fingerprint.

Cite this