SVMTool: A general POS Tagger Generator Based on Support Vector Machines
By Jesús Giménez, Lluís Màrquez i Villodre. 2004
422 citations recorded by the source corpus.
This paper presents the SVMTool, a simple, flexible, effective and efficient part–of–speech tagger based on Support Vector Machines. The SVMTool offers a fairly good balance among these properties which make it really practical for current NLP applications. It is very easy to use and easily configurable so as to perfectly fit the needs of a number of different applications. Results are also very competitive, achieving an accuracy of 97.16% for English on the Wall Street Journal corpus. It has been also successfully applied to Spanish exhibiting a similar performance. A first release of the SVMTool Perl prototype is now freely available for public use. A most efficient C++ version is coming very soon.
Subfields: Artificial Intelligence, Natural Language Processing, Information Extraction
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Mohit Bansal
- Noah A. Smith
- Christopher D. Manning
- Yoav Goldberg
- Yoshua Bengio
- Graham Neubig
- Yonatan Belinkov
- Luke S. Zettlemoyer
In Natural Language Processing
- Yoav Goldberg
- Christopher D. Manning
- Hinrich Schütze
- Yoshua Bengio
- Yonatan Belinkov
- Jason Aaron Edward Weston
- Graham Neubig
- Mohit Bansal
In Information Extraction
- Richard Johansson
- Paola Merlo
- Gabriele Antonio Musillo
- Anders Søgaard
- Jason Aaron Edward Weston
- Jörg Tiedemann
- Richard Sproat
- Navdeep Jaitly
Citations in the corpus
- Natural Language Processing (Almost) from Scratch (2011, methodology): “Giménez and Màrquez (2004) proposed a SVM approach also trained on text windows, with bidirectional inference achieved with two Viterbi decoders (left-to-right and right-to-left).”
- Parallel Data, Tools and Interfaces in OPUS (2012, background): “1 models, “zpar” refers to a statistical parser with language-specific features for Chinese and English (Zhang and Nivre, 2011), SVMTool (Giménez and Màrquez, 2004) and MElt (Denis and Sagot, 2009) are statistical taggers with language-specific models required for the respective dependency parser mo”
- Search-based structured prediction (2009, neutral): “The second variety is typified by maximum entropy Markov models [37], though the basic idea of MEMMs has also been applied more generally to SVMs [27,28,20].”
- MADA + TOKAN : A Toolkit for Arabic Tokenization , Diacritization , Morphological Disambiguation , POS Tagging , Stemming and Lemmatization (2009, background): “]=<gloss> Once MADA ranks the possible analyses, it appends a numerical score to each analysis, and flags the topscoring analysis for each word in each context.”
- IXA pipeline: Efficient and Ready to Use Multilingual NLP tools (2014, background): “Gim´enez and Marquez (2004) report 98.86%, although they train and test on a different sub-set of the Ancora corpus.”
- We Need to Talk about Standard Splits (2019, methodology): “Three models—SVMTool (Giménez and Màrquez, 2004), MElt (Denis and Sagot, 2009), and Morče/COMPOST (Spoustová et al., 2009)— produced substantial compilation or runtime errors.”
- Efficient Higher-Order CRFs for Morphological Tagging (2013, methodology): “We use the following baselines: SVMTool (Giménez and Màrquez, 2004), an SVM-based discriminative tagger; RFTagger (Schmid and Laws, 2008), an n-gram Hidden Markov Model (HMM) tagger developed for POS+MORPH tagging; Morfette (Chrupała et al.”
- RNN Approaches to Text Normalization: A Challenge (2016, methodology): “2The morphosyntactic tagger is an SVM model using hand-tuned features that classify the morphological bundle for each word independently, similar to SVMTool (Giménez and Màrquez, 2004) and MateTagger (Bohnet and Nivre, 2012).”
- Robust Parsing of the Proposition Bank (2006, methodology): “To provide the augmented model with tagged input sentences, we trained an SVM tagger whose features and parameters are described in detail in (Gimenez and Marquez, 2004).”
- Guided Learning for Bidirectional Sequence Classification (2007, background): “85 (Giménez and Màrquez, 2004) - 2.”
- Semi-supervised condensed nearest neighbor for part-of-speech tagging (2011, methodology): “…the standard splits and construct our data set in the following way, following Søgaard (2010): Each word in the data w i is associated with a feature vector where x i is the prediction on w i of a supervised part-of-speech tagger, in our case SVMTool 1 (Gimenez and Marquez, 2004) trained on Sect.”
- The CoNLL 2008 Shared Task on Joint Parsing of Syntactic and Semantic Dependencies (2008, methodology): “The PPOS and PPOSS fields were automatically predicted using the SVMTool POS tagger (Gim´enez, 2004).”