===== Tamil (ta) =====

[[http://ufal.mff.cuni.cz/~ramasamy/tamiltb/0.1/|Tamil Dependency Treebank]] (TamilTB)

==== Versions ====

  * TamilTB 0.1

==== Obtaining and License ====

TamilTB 0.1 is [[http://ufal.mff.cuni.cz/~ramasamy/tamiltb/0.1/download.html|distributed]] under the [[http://creativecommons.org/licenses/by-nc-sa/3.0/|Creative Commons by-nc-sa license]]. The license in short:

  * non-commercial usage
  * redistribution permitted
  * attribution to Charles University in Prague, Institute of Formal and Applied Linguistics required
    * cite one of the principal publications (see below) in published work using the treebank

TamilTB was created by members of the [[http://ufal.mff.cuni.cz/|Institute of Formal and Applied Linguistics]] (Ústav formální a aplikované lingvistiky, ÚFAL), Faculty of Mathematics and Physics (Matematicko-fyzikální fakulta), Charles University in Prague (Univerzita Karlova v Praze), Malostranské náměstí 25, Praha, CZ-11800, Czechia.

==== References ====

  * Website
    * http://ufal.mff.cuni.cz/~ramasamy/tamiltb/0.1/
  * Data
    * //no separate citation//
  * Principal publications
    * Loganathan Ramasamy, Zdeněk Žabokrtský: [[http://www.springerlink.com/content/w18v7621070h51g1/|Tamil Dependency Parsing: Results using Rule Based and Corpus Based Approaches]]. In: //Proceedings of the 12th International Conference on Computational Linguistics and Intelligent Text Processing (CICLing 2011) – Volume Part I//, pages 82-95, Tokyo, Japan, 2011, published by Springer Berlin / Heidelberg, ISBN 978-3-642-19399-6.
    * Loganathan Ramasamy, Zdeněk Žabokrtský: Prague Dependency Style Treebank for Tamil. In: //Proceedings of the 8th International Conference on Language Resources and Evaluation (LREC 2012)//, İstanbul, Turkey, 2012
  * Documentation
    * [[http://ufal.mff.cuni.cz/~ramasamy/tamiltb/0.1/morph_annotation.html|Morphological annotation]]
    * [[http://ufal.mff.cuni.cz/~ramasamy/tamiltb/0.1/dependency_annotation.html|Syntactic annotation]]
    * Loganathan Ramasamy, Zdeněk Žabokrtský: [[http://ufal.mff.cuni.cz/~ramasamy/papers/2011-TamilTB-TR.pdf|Tamil Dependency Treebank (TamilTB) – 0.1 Annotation Manual]]. Technical Report TR-2011-42, ÚFAL MFF UK, Praha, Czechia, 2011

==== Domain ====

News (http://www.dinamani.com/).

==== Size ====

Version 0.1 contains 9581 tokens in 600 sentences, yielding 15.97 tokens per sentence on average. We defined the following data split: 7592 tokens / 480 sentences training, 1989 tokens / 120 sentences test.

==== Inside ====

Tamil script has been [[http://ufal.mff.cuni.cz/~ramasamy/tamiltb/0.1/introduction.html#Text_preprocessing|romanized]] (the romanization is case-sensitive).

The treebank is distributed in three formats: TMT ([[http://ufal.mff.cuni.cz/tectomt/|TectoMT]] XML), [[:formát CoNLL|CoNLL]] and TnT-tagger style (only POS-tagged layer).

Morphological annotation is manual and it includes lemmas, parts of speech and morphosyntactic features. Syntactic annotation follows the style of the [[cs|Prague Dependency Treebank]].

==== Sample ====

The first sentence of the CoNLL version of training data:

| 1 | cennai | cennai | N | <nowiki>NEN-3SN--</nowiki> | <nowiki>Cas=N|Per=3|Num=S|Gen=N</nowiki> | 2 | AAdjn | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 2 | arukE | arukE | P | <nowiki>PP-------</nowiki> | <nowiki>_</nowiki> | 18 | AuxP | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 3 | sri | sri | N | <nowiki>NEN-3SN--</nowiki> | <nowiki>Cas=N|Per=3|Num=S|Gen=N</nowiki> | 4 | Atr | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 4 | perumpuTUril | perumpuTUr | N | <nowiki>NEL-3SN--</nowiki> | <nowiki>Cas=L|Per=3|Num=S|Gen=N</nowiki> | 18 | AAdjn | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 5 | kirIn | kirIn | N | <nowiki>NEN-3SN--</nowiki> | <nowiki>Cas=N|Per=3|Num=S|Gen=N</nowiki> | 6 | Atr | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 6 | pIltu | pIltu | N | <nowiki>NEN-3SN--</nowiki> | <nowiki>Cas=N|Per=3|Num=S|Gen=N</nowiki> | 11 | Atr | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 7 | <nowiki>(</nowiki> | <nowiki>(</nowiki> | Z | <nowiki>Z:-------</nowiki> | <nowiki>_</nowiki> | 6 | AuxG | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 8 | wavIna | wavInam | J | <nowiki>JJ-------</nowiki> | <nowiki>_</nowiki> | 6 | Atr | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 9 | <nowiki>)</nowiki> | <nowiki>)</nowiki> | Z | <nowiki>Z:-------</nowiki> | <nowiki>_</nowiki> | 6 | AuxG | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 10 | vimAna | vimAnam | N | <nowiki>NO--3SN--</nowiki> | <nowiki>Per=3|Num=S|Gen=N</nowiki> | 11 | Atr | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 11 | wilaiyaTTukkukk | wilaiyam | N | <nowiki>NND-3SN--</nowiki> | <nowiki>Cas=D|Per=3|Num=S|Gen=N</nowiki> | 12 | Atr | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 12 | Ana | Aku | T | <nowiki>Tg-------</nowiki> | <nowiki>_</nowiki> | 13 | Atr | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 13 | wilam | wilam | N | <nowiki>NNN-3SN--</nowiki> | <nowiki>Cas=N|Per=3|Num=S|Gen=N</nowiki> | 18 | Sb | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 14 | yArukkum | yAr | R | <nowiki>RBD-3SA--</nowiki> | <nowiki>Cas=D|Per=3|Num=S|Gen=A</nowiki> | 15 | Atr | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 15 | pATippu | pATippu | N | <nowiki>NNN-3SN--</nowiki> | <nowiki>Cas=N|Per=3|Num=S|Gen=N</nowiki> | 16 | Comp | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 16 | illATa | il | P | <nowiki>PP-------</nowiki> | <nowiki>_</nowiki> | 17 | AuxP | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 17 | vakaiyil | vakai | N | <nowiki>NNL-3SN--</nowiki> | <nowiki>Cas=L|Per=3|Num=S|Gen=N</nowiki> | 18 | AAdjn | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 18 | etukkap | etu | V | <nowiki>Vu-T---AA</nowiki> | <nowiki>Ten=T|Voi=A|Neg=A</nowiki> | 20 | Obj | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 19 | patum | patu | V | <nowiki>VR-F3SNPA</nowiki> | <nowiki>Ten=F|Per=3|Num=S|Gen=N|Voi=P|Neg=A</nowiki> | 18 | AuxV | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 20 | enRu | en | T | <nowiki>Tt-T----A</nowiki> | <nowiki>Ten=T|Neg=A</nowiki> | 23 | AuxC | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 21 | muTalvar | muTalvar | N | <nowiki>NNN-3SH--</nowiki> | <nowiki>Cas=N|Per=3|Num=S|Gen=H</nowiki> | 22 | Atr | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 22 | karuNAwiTi | karuNAwiTi | N | <nowiki>NEN-3SH--</nowiki> | <nowiki>Cas=N|Per=3|Num=S|Gen=H</nowiki> | 23 | Sb | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 23 | uRuTiyaLiTT | uRuTiyaLi | V | <nowiki>Vt-T---AA</nowiki> | <nowiki>Ten=T|Voi=A|Neg=A</nowiki> | 0 | Pred | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 24 | uLLAr | uL | V | <nowiki>VR-T3SHAA</nowiki> | <nowiki>Ten=T|Per=3|Num=S|Gen=H|Voi=A|Neg=A</nowiki> | 23 | AuxV | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 25 | <nowiki>.</nowiki> | <nowiki>.</nowiki> | Z | <nowiki>Z#-------</nowiki> | <nowiki>_</nowiki> | 0 | AuxK | <nowiki>_</nowiki> | <nowiki>_</nowiki> |

The first sentence of the CoNLL version of test data:

| 1 | pikAr | pikAr | N | <nowiki>NEN-3SN--</nowiki> | <nowiki>Cas=N|Per=3|Num=S|Gen=N</nowiki> | 2 | Atr | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 2 | iliruwTu | iliruwTu | P | <nowiki>PP-------</nowiki> | <nowiki>_</nowiki> | 4 | AuxP | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 3 | ErALamAna | ErALamAna | J | <nowiki>JJ-------</nowiki> | <nowiki>_</nowiki> | 4 | Atr | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 4 | iLainjarkaL | iLainjar | N | <nowiki>NNN-3PA--</nowiki> | <nowiki>Cas=N|Per=3|Num=P|Gen=A</nowiki> | 9 | Sb | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 5 | vElai | vElai | N | <nowiki>NNN-3SN--</nowiki> | <nowiki>Cas=N|Per=3|Num=S|Gen=N</nowiki> | 6 | Obj | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 6 | TEti | TEtu | V | <nowiki>Vt-T---AA</nowiki> | <nowiki>Ten=T|Voi=A|Neg=A</nowiki> | 9 | AAdjn | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 7 | veLi | veLi | J | <nowiki>JJ-------</nowiki> | <nowiki>_</nowiki> | 8 | Atr | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 8 | mAwilangkaLukku | mAwilam | N | <nowiki>NND-3PN--</nowiki> | <nowiki>Cas=D|Per=3|Num=P|Gen=N</nowiki> | 9 | AAdjn | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 9 | kutipeyarwTu | kutipeyar | V | <nowiki>Vt-T---AA</nowiki> | <nowiki>Ten=T|Voi=A|Neg=A</nowiki> | 0 | Pred | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 10 | varukinRanar | varu | V | <nowiki>VR-P3PHAA</nowiki> | <nowiki>Ten=P|Per=3|Num=P|Gen=H|Voi=A|Neg=A</nowiki> | 9 | AuxV | <nowiki>_</nowiki> | <nowiki>_</nowiki> |
| 11 | <nowiki>.</nowiki> | <nowiki>.</nowiki> | Z | <nowiki>Z#-------</nowiki> | <nowiki>_</nowiki> | 0 | AuxK | <nowiki>_</nowiki> | <nowiki>_</nowiki> |

==== Parsing ====

Nonprojectivities in PADT are very rare. Only 15 of the 9581 tokens are attached nonprojectively (0.16%).

Initial parsing results were published by [[http://ufal.mff.cuni.cz/~ramasamy/papers/2011-pres-CICLing.pdf|(Ramasamy and Žabokrtský, 2011)]]. They use smaller data and different training-test data split than defined here (2008 tokens training, 953 tokens test).

^ Parser (Authors) ^ LAS ^ UAS ^
| Malt (Nivre et al.) | 65.69 | 75.03 |
| MST (McDonald et al.) | 65.69 | 74.92 |