Linear Co-occurrence Rate Networks (L-CRNs) for Sequence Labeling

Zhemin Zhu, Djoerd Hiemstra, Peter M.G. Apers

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

1 Citation (Scopus)
8 Downloads (Pure)

Abstract

Sequence labeling has wide applications in natural language processing and speech processing. Popular sequence labeling models suffer from some known problems. Hidden Markov models (HMMs) are generative models and they cannot encode transition features; Conditional Markov models (CMMs) suffer from the label bias problem; And training of conditional random fields (CRFs) can be expensive. In this paper, we propose Linear Co-occurrence Rate Networks (L-CRNs) for sequence labeling which avoid the mentioned problems with existing models. The factors of L-CRNs can be locally normalized and trained separately, which leads to a simple and efficient training method. Experimental results on real-world natural language processing data sets show that L-CRNs reduce the training time by orders of magnitudes while achieve very competitive results to CRFs.
Original languageUndefined
Title of host publicationProceedings of the Second International Conference on Statistical Language and Speech Processing, SLSP 2014
Place of PublicationBerlin
PublisherSpringer
Pages185-196
Number of pages12
ISBN (Print)978-3-319-11396-8
DOIs
Publication statusPublished - Oct 2014

Publication series

NameLecture Notes in Artifical Intelligence
PublisherSpringer Verlag
Volume8791
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Keywords

  • EWI-25353
  • DB-DM: DATA MINING
  • HMMs
  • METIS-309684
  • CRFs
  • IR-93325
  • Co-occurrence rate
  • Sequence labeling

Cite this

Zhu, Z., Hiemstra, D., & Apers, P. M. G. (2014). Linear Co-occurrence Rate Networks (L-CRNs) for Sequence Labeling. In Proceedings of the Second International Conference on Statistical Language and Speech Processing, SLSP 2014 (pp. 185-196). (Lecture Notes in Artifical Intelligence; Vol. 8791). Berlin: Springer. https://doi.org/10.1007/978-3-319-11397-5_14
Zhu, Zhemin ; Hiemstra, Djoerd ; Apers, Peter M.G. / Linear Co-occurrence Rate Networks (L-CRNs) for Sequence Labeling. Proceedings of the Second International Conference on Statistical Language and Speech Processing, SLSP 2014. Berlin : Springer, 2014. pp. 185-196 (Lecture Notes in Artifical Intelligence).
@inproceedings{de4a054b4aed4ef2b45041a2be37563a,
title = "Linear Co-occurrence Rate Networks (L-CRNs) for Sequence Labeling",
abstract = "Sequence labeling has wide applications in natural language processing and speech processing. Popular sequence labeling models suffer from some known problems. Hidden Markov models (HMMs) are generative models and they cannot encode transition features; Conditional Markov models (CMMs) suffer from the label bias problem; And training of conditional random fields (CRFs) can be expensive. In this paper, we propose Linear Co-occurrence Rate Networks (L-CRNs) for sequence labeling which avoid the mentioned problems with existing models. The factors of L-CRNs can be locally normalized and trained separately, which leads to a simple and efficient training method. Experimental results on real-world natural language processing data sets show that L-CRNs reduce the training time by orders of magnitudes while achieve very competitive results to CRFs.",
keywords = "EWI-25353, DB-DM: DATA MINING, HMMs, METIS-309684, CRFs, IR-93325, Co-occurrence rate, Sequence labeling",
author = "Zhemin Zhu and Djoerd Hiemstra and Apers, {Peter M.G.}",
note = "eemcs-eprint-25353",
year = "2014",
month = "10",
doi = "10.1007/978-3-319-11397-5_14",
language = "Undefined",
isbn = "978-3-319-11396-8",
series = "Lecture Notes in Artifical Intelligence",
publisher = "Springer",
pages = "185--196",
booktitle = "Proceedings of the Second International Conference on Statistical Language and Speech Processing, SLSP 2014",

}

Zhu, Z, Hiemstra, D & Apers, PMG 2014, Linear Co-occurrence Rate Networks (L-CRNs) for Sequence Labeling. in Proceedings of the Second International Conference on Statistical Language and Speech Processing, SLSP 2014. Lecture Notes in Artifical Intelligence, vol. 8791, Springer, Berlin, pp. 185-196. https://doi.org/10.1007/978-3-319-11397-5_14

Linear Co-occurrence Rate Networks (L-CRNs) for Sequence Labeling. / Zhu, Zhemin; Hiemstra, Djoerd; Apers, Peter M.G.

Proceedings of the Second International Conference on Statistical Language and Speech Processing, SLSP 2014. Berlin : Springer, 2014. p. 185-196 (Lecture Notes in Artifical Intelligence; Vol. 8791).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › Academic › peer-review

TY - GEN

T1 - Linear Co-occurrence Rate Networks (L-CRNs) for Sequence Labeling

AU - Zhu, Zhemin

AU - Hiemstra, Djoerd

AU - Apers, Peter M.G.

N1 - eemcs-eprint-25353

PY - 2014/10

Y1 - 2014/10

N2 - Sequence labeling has wide applications in natural language processing and speech processing. Popular sequence labeling models suffer from some known problems. Hidden Markov models (HMMs) are generative models and they cannot encode transition features; Conditional Markov models (CMMs) suffer from the label bias problem; And training of conditional random fields (CRFs) can be expensive. In this paper, we propose Linear Co-occurrence Rate Networks (L-CRNs) for sequence labeling which avoid the mentioned problems with existing models. The factors of L-CRNs can be locally normalized and trained separately, which leads to a simple and efficient training method. Experimental results on real-world natural language processing data sets show that L-CRNs reduce the training time by orders of magnitudes while achieve very competitive results to CRFs.

AB - Sequence labeling has wide applications in natural language processing and speech processing. Popular sequence labeling models suffer from some known problems. Hidden Markov models (HMMs) are generative models and they cannot encode transition features; Conditional Markov models (CMMs) suffer from the label bias problem; And training of conditional random fields (CRFs) can be expensive. In this paper, we propose Linear Co-occurrence Rate Networks (L-CRNs) for sequence labeling which avoid the mentioned problems with existing models. The factors of L-CRNs can be locally normalized and trained separately, which leads to a simple and efficient training method. Experimental results on real-world natural language processing data sets show that L-CRNs reduce the training time by orders of magnitudes while achieve very competitive results to CRFs.

KW - EWI-25353

KW - DB-DM: DATA MINING

KW - HMMs

KW - METIS-309684

KW - CRFs

KW - IR-93325

KW - Co-occurrence rate

KW - Sequence labeling

U2 - 10.1007/978-3-319-11397-5_14

DO - 10.1007/978-3-319-11397-5_14

M3 - Conference contribution

SN - 978-3-319-11396-8

T3 - Lecture Notes in Artifical Intelligence

SP - 185

EP - 196

BT - Proceedings of the Second International Conference on Statistical Language and Speech Processing, SLSP 2014

PB - Springer

CY - Berlin

ER -

Zhu Z, Hiemstra D, Apers PMG. Linear Co-occurrence Rate Networks (L-CRNs) for Sequence Labeling. In Proceedings of the Second International Conference on Statistical Language and Speech Processing, SLSP 2014. Berlin: Springer. 2014. p. 185-196. (Lecture Notes in Artifical Intelligence). https://doi.org/10.1007/978-3-319-11397-5_14

Access to Document