Skip to main navigation Skip to search Skip to main content

Iterative multiple sequence labeling with classifier combination

  • Xinxin Li*
  • , Xuan Wang
  • , Lin Yao
  • *Corresponding author for this work
  • Harbin Institute of Technology Shenzhen

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

Abstract

Traditional pipeline approach causes error propagation and cannot share information among multiple tasks. In this paper, we proposed an iterative approach for sequence labeling problems with classifier combination. The approach is beneficial for both cascaded tasks and multiple separate tasks. We discuss feature selection strategy to increase diversity and obtain better oracle for classifier combination. An averaged perceptron algorithm is used as the strategy of classifier combination. Experimental results on POS tagging and chunking problem show that our approach outperforms pipeline, tag combination, and other classifier combination approaches.

Original languageEnglish
Title of host publicationNLP-KE 2011 - Proceedings of the 7th International Conference on Natural Language Processing and Knowledge Engineering
Pages397-400
Number of pages4
DOIs
StatePublished - 2011
Externally publishedYes
Event7th International Conference on Natural Language Processing and Knowledge Engineering, NLP-KE 2011 - Tokushima, Japan
Duration: 27 Nov 201129 Nov 2011

Publication series

NameNLP-KE 2011 - Proceedings of the 7th International Conference on Natural Language Processing and Knowledge Engineering

Conference

Conference7th International Conference on Natural Language Processing and Knowledge Engineering, NLP-KE 2011
Country/TerritoryJapan
CityTokushima
Period27/11/1129/11/11

Keywords

  • averaged perceptron
  • classifier combination
  • iterative approach
  • sequence labeling

Fingerprint

Dive into the research topics of 'Iterative multiple sequence labeling with classifier combination'. Together they form a unique fingerprint.

Cite this