---
title: "SVMTool: A general POS Tagger Generator Based — Giménez 2004"
description: "The paper presents SVMTool, a toolkit for building part-of-speech taggers."
authors: ["Jesús Giménez", "Lluís Màrquez"]
year: 2004
venue: "Proceedings of the Fourth International Conference on Language Resources and Evaluation (LREC'04)"
source: "https://aclanthology.org/L04-1373/"
licence: "CC BY-NC 4.0 (ELRA, LREC proceedings)"
licence_url: "https://creativecommons.org/licenses/by-nc/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/papers/gimenez-2004-svmtool-general-pos-tagger-generator/"
---
[Elamur Research](https://md.research.elamur.ai/) · [this page in HTML](https://research.elamur.ai/free/papers/gimenez-2004-svmtool-general-pos-tagger-generator/)

[home](https://md.research.elamur.ai/) / [papers](https://md.research.elamur.ai/free/papers/) / [2004](https://md.research.elamur.ai/free/papers/2004/) / SVMTool: A general POS Tagger Generator Based on Support Vector Machines

# SVMTool: A general POS Tagger Generator Based on Support Vector Machines

**Authors:** [Jesús Giménez](https://md.research.elamur.ai/free/authors/jesus-gimenez/), [Lluís Màrquez](https://md.research.elamur.ai/free/authors/lluis-marquez/) · **Year:** 2004 · **Venue:** Proceedings of the Fourth International Conference on Language Resources and Evaluation (LREC'04) · **Licence:** [CC BY-NC 4.0 (ELRA, LREC proceedings)](https://creativecommons.org/licenses/by-nc/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/L04-1373/) · [PDF](https://aclanthology.org/L04-1373.pdf) · [OpenAlex](https://openalex.org/W2399720833)

**Citations:** 339 citations (openalex-snapshot · 2026-10-14) · [112](#cited-by) in our library, of the 140673 papers whose bibliographies we read

## Summary
The paper presents SVMTool, a toolkit for building part-of-speech taggers. It aims to balance accuracy, speed, ease of use and language independence. Each tagging decision is made by linear Support Vector Machine classifiers trained with SVM-light in the primal formulation. They draw on a large, configurable set of contextual features: word and POS n-grams, ambiguity classes, "may be" indicators, and affix and orthographic cues for unknown words. The package has three parts: a learner (SVMTlearn), a tagger (SVMTagger) and an evaluator (SVMTeval), plus an API for embedded use. Users can choose greedy or Viterbi-based sentence-level decoding, the tagging direction, one or two passes, and a backup morphological lexicon. On the Wall Street Journal corpus the tool reaches 97.16% accuracy. On the Spanish LEXESP corpus it reaches 96.89%, and both scores beat TnT under identical conditions. The Perl prototype tags about 1,500 words per second. The work matters because it offers a freely available tagger generator that is competitive, flexible and adaptable to new languages using only an annotated corpus.

## Key points
- **English setup and results:** WSJ sections 0–18 were used for training (912K words), 19–21 for validation and 22–24 for testing. SVMTool scored 97.16% overall versus 96.46% for TnT, and 89.01% versus 85.86% on unknown words. Unknown words made up 2.81% of the test set.
- **Spanish setup and results:** LEXESP (106K words) was split into 86K words for training and 20K for testing, with 12.21% unknown words. SVMTool scored 96.89% versus 96.50% for TnT. When a morphological analyzer supplied a backup lexicon, the scores rose to 98.86% versus 98.41%.
- **Efficiency:** Tagging runs at about 1,500 words per second on a Pentium IV at 2GHz. The heaviest English training run, using the full feature set and fine C tuning, took about 24 CPU hours.
- **Model compression:** Removing small weight components up to about 70% of all weights left accuracy unchanged. Accuracy dropped below the state-of-the-art range (97.0–97.2%) only after about 95% had been removed.
- **Default settings:** The feature window is 5 tokens with the target word at position 2. The feature space is capped at 100,000 dimensions, and features seen only once are discarded.

## Interesting ideas
- **Bidirectional combination:** The tagger can run left-to-right and right-to-left and then combine both outputs, which gave a significant gain at little cost. This is a cheap trick for any greedy sequence labeller where the processing order biases errors.
- **Pruning linear model weights:** Most of the weight components can be zeroed out with little loss in accuracy. This is a simple way to shrink and speed up linear models before deploying them.
- **Ambiguity-class features:** The tagger encodes which tags a word could take ("may be" features), not just which tags were already assigned. This lexicon-derived feature transfers well to morphologically rich languages, where dictionaries can narrow the candidate tags.
- **Breaking down evaluation results:** SVMTeval reports accuracy separately for known and unknown words, for ambiguous words, and by ambiguity class or class size. This shows more clearly where a tagger fails than a single overall number does.
- **Exposing scores for all candidate tags:** The API returns the score for every candidate tag, not only the winning one. Downstream components can then handle near-tied decisions themselves, for example by flagging them for review or passing several candidates to a parser.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

Fields: [Linguistics & languages](https://md.research.elamur.ai/free/shelves/linguistics/)

Topics: [Part-of-speech tagging](https://md.research.elamur.ai/free/topics/part-of-speech-tagging/)

## Cited by

Cited by

112 papers of our library cite it (of the 140673 whose bibliographies we read); 10 of them have a page here, 102 not yet.

- [Arabic Diacritization through Full Morphological Tagging](https://md.research.elamur.ai/free/papers/aclbulk-2007-habash-arabic-diacritization-through-full-morphological/)
- [Evaluation of Dependency Parsers on Unbounded Dependencies](https://md.research.elamur.ai/free/papers/aclbulk-2010-nivre-evaluation-dependency-parsers-unbounded-dependencies/)
- [Part-of-Speech Tagging for Historical English](https://md.research.elamur.ai/free/papers/aclbulk-2016-yang-part-speech-tagging-historical-english/)
- [Natural Language Processing (almost) from Scratch](https://md.research.elamur.ai/free/papers/collobert-2011-natural-language-processing-almost-scratch/)
- [Bidirectional LSTM-CRF Models for Sequence Tagging](https://md.research.elamur.ai/free/papers/huang-2015-bidirectional-lstm-crf-models-sequence/)
- [End-to-end Sequence Labeling via Bi-directional LSTM-CNNs-CRF](https://md.research.elamur.ai/free/papers/ma-2016-end-end-sequence-labeling-bi/)
- [Exploiting Semantic Role Labeling, WordNet and Wikipedia for Coreference Resolution](https://md.research.elamur.ai/free/papers/ponzetto-2006-exploiting-semantic-role-labeling-wordnet/)
- [A Global Relaxation Labeling Approach to Coreference Resolution](https://md.research.elamur.ai/free/papers/sapena-2010-global-relaxation-labeling-approach-coreference/)
- [The CoNLL 2008 Shared Task on Joint Parsing of Syntactic and Semantic Dependencies](https://md.research.elamur.ai/free/papers/surdeanu-2008-conll-2008-shared-task-joint/)
- [Application-driven Statistical Paraphrase Generation](https://md.research.elamur.ai/free/papers/zhao-2009-application-driven-statistical-paraphrase-generation/)

## Cite

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{gimenez2004svmtool,
  title = {{SVMTool: A general POS Tagger Generator Based on Support Vector Machines}},
  author = {Giménez, Jesús and Màrquez, Lluís},
  booktitle = {Proceedings of the Fourth International Conference on Language Resources and Evaluation (LREC'04)},
  year = {2004},
  doi = {10.63317/52pkqry5zu2h},
  url = {https://doi.org/10.63317/52pkqry5zu2h}
}
```

- [← previous Statistical Machine Translation with Scarce Resources Using Morpho-syntactic Information](https://md.research.elamur.ai/free/papers/en-2004-statistical-machine-translation-scarce-resources/)
- [next → Synchronous Grammars as Tree Transducers](https://md.research.elamur.ai/free/papers/shieber-2004-synchronous-grammars-as-tree-transducers/)
- [↑ back to papers of 2004](https://md.research.elamur.ai/free/papers/2004/)

## References

8 references, 3 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Brants, T., 2000. TnT - A Statistical Part-of-Speech Tagger. In Proceedings of the Sixth ANLP. [find](https://scholar.google.com/scholar?q=Brants%2C+T%2E%2C+2000%2E+TnT+%2D+A+Statistical+Part%2Dof%2DSpeech+Tagger%2E+In+Proceedings+of+the+Sixth+ANLP%2E) [our card](https://md.research.elamur.ai/free/papers/brants-2000-tnt-statistical-part-speech-tagger/)
2. Collins, M., 2002. Discriminative Training Methods for Hidden Markov Models: Theory and Experiments with Perceptron Algorithms. In Proceedings of the 7th EMNLP Conference. [find](https://scholar.google.com/scholar?q=Collins%2C+M%2E%2C+2002%2E+Discriminative+Training+Methods+for+Hidden+Markov+Models%3A+Theory+and+Experiments+with+Perceptron+Algorithms%2E+In+Proceedings+of+the+7th+EMNLP+Conference%2E) [our card](https://md.research.elamur.ai/free/papers/collins-2002-discriminative-training-methods-hidden-markov/)
3. Cristianini, N. and J. Shawe-Taylor, 2000. An Introduction to Support Vector Machines. Cambridge University Press. [find](https://scholar.google.com/scholar?q=Cristianini%2C+N%2E+and+J%2E+Shawe%2DTaylor%2C+2000%2E+An+Introduction+to+Support+Vector+Machines%2E+Cambridge+University+Press%2E)
4. Giménez, J. and L. Màrquez, 2003. Fast and Accurate Part-of-Speech Tagging: The SVM Approach Revisited. In Proceedings of the Fourth RANLP. [find](https://scholar.google.com/scholar?q=Gime%CC%81nez%2C+J%2E+and+L%2E+Ma%CC%80rquez%2C+2003%2E+Fast+and+Accurate+Part%2Dof%2DSpeech+Tagging%3A+The+SVM+Approach+Revisited%2E+In+Proceedings+of+the+Fourth+RANLP%2E)
5. Joachims, T., 1999. Making large-Scale SVM Learning Practical. MIT-Press. [find](https://scholar.google.com/scholar?q=Joachims%2C+T%2E%2C+1999%2E+Making+large%2DScale+SVM+Learning+Practical%2E+MIT%2DPress%2E)
6. Mihalcea, Rada, 2003. The Role of Non-Ambiguous Words in Natural Language Disambiguation. In Proceedings of the Fourth RANLP. [find](https://scholar.google.com/scholar?q=Mihalcea%2C+Rada%2C+2003%2E+The+Role+of+Non%2DAmbiguous+Words+in+Natural+Language+Disambiguation%2E+In+Proceedings+of+the+Fourth+RANLP%2E)
7. Toutanova, K., D. Klein, and C. D. Manning, 2003. Feature-rich part-of-speech tagging with a cyclic dependency network. In Proceedings of HLT-NAACL’03. [find](https://scholar.google.com/scholar?q=Toutanova%2C+K%2E%2C+D%2E+Klein%2C+and+C%2E+D%2E+Manning%2C+2003%2E+Feature%2Drich+part%2Dof%2Dspeech+tagging+with+a+cyclic+dependency+network%2E+In+Proceedings+of+HLT%2DNAACL%E2%80%9903%2E)
8. X.Carreras, I. Chao, L. Padró, and M. Padró, 2004. Freeling: An open-source suite of language analyzers. In Proceedings of the 4th LREC Conference. [find](https://scholar.google.com/scholar?q=X%2ECarreras%2C+I%2E+Chao%2C+L%2E+Padro%CC%81%2C+and+M%2E+Padro%CC%81%2C+2004%2E+Freeling%3A+An+open%2Dsource+suite+of+language+analyzers%2E+In+Proceedings+of+the+4th+LREC+Conference%2E) [our card](https://md.research.elamur.ai/free/papers/carreras-2004-freeling-open-source-suite-language/)