Tips and Tricks of the Prague Arabic Dependency Treebank

What it is about

A practical guide from the team that builds the Prague Arabic Dependency Treebank (PADT), which annotates MSA at three levels: morphology, surface syntax (analytical trees with functions such as Pred, Sb, Obj, Atr, Coord) and a tectogrammatical level. The paper shows how the tasks are done in code: Unicode-class regular expressions that separate Arabic words, Latin words, numbers and punctuation; mode-based conversion between orthography, Buckwalter transliteration and ArabTeX; a Haskell lexicon with typed records; the MorphoTrees method, which lays out the readings of a word as a hierarchy that an annotator narrows by feature; and the tools around the treebank (TrEd, Netgraph search, version control and consistency checks). It reports no experiments and no numbers; it is a list of design choices and their reasons.

Ideas

• I1. Token classes by Unicode range: the paper's regular expressions split a text into Arabic words (including the diacritic and tatweel ranges), Latin words, digit groups and punctuation. The same kind of table, with the Hebrew cantillation and niqqud ranges and the maqaf and sof pasuq, can be checked against tools/langseg and the tokenizer of tools/hbo, with a test on a verse with full cantillation.
• I2. Store the lexicon in a notation between orthography and phonology, and render orthographic forms from it on demand. The paper's reason is that the model is then written once and applies to unpointed and pointed text alike. This is the design that tools/hbo needs for its pointed and unpointed modes, and it fits the abstract notation idea in the Mova plan for The lexicon.
• I3. Coordination as its own node: the paper marks a Coord node, marks each conjunct as a member, and attaches the modifiers of the whole coordination to the Coord node. Check how the Hebrew trees from PTNK handle coordinated subjects and objects, and write a small test set of coordinated clauses from PTNK with the expected structure.

For Mova

I1 is a check for tools/langseg and tools/hbo (tokenizer) and for 2. Cantillation: the accents (te'amim). I2 is a design rule for the lexicon in The lexicon and the modes of tools/hbo. I3 is a test for treebanks/hbo/ptnk and a point for Syntax: the clause and the sentence.

Topics: Treebanks

Cite

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

@inproceedings{smrz2006tips,
  title = {{Tips and Tricks of the Prague Arabic Dependency Treebank}},
  author = {Smrž, Otakar},
  booktitle = {Proceedings of the International Conference on the Challenge of Arabic for NLP/MT},
  year = {2006},
  url = {https://aclanthology.org/2006.bcs-1.3/}
}

References

20 references, 4 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Buckwalter, Tim (2002). Buckwalter Arabic Morphological Analyzer 1.0. LDC catalog number LDC2002L49, ISBN 1-58563-257-0. find
2. Buckwalter, Tim (2004). Buckwalter Arabic Morphological Analyzer 2.0. LDC catalog number LDC2004L02, ISBN 1-58563-324-0. find
3. Chiang, David and Mona Diab and Nizar Habash and Owen Rambow and Safiullah Sharif (2006). ‘Parsing Arabic Dialects’. In Proceedings of the 11th Conference of the European Chapter of the Association for Computational Linguistics, pages 369–376, Trento, Italy. find our card
4. Corston-Oliver, Simon and Anthony Aue and Kevin Duh and Eric Ringger (2006). ‘Multilingual Dependency Parsing using Bayes Point Machines’. In Proceedings of HLT-NAACL 2006, pages 160–167, New York. find
5. El Dada, Ali and Aarne Ranta (2006). ‘Implementing an Open Source Arabic Resource Grammar in Grammatical Framework’. In Proceedings of the XXth Arabic Linguistics Symposium. John Benjamins. find
6. Debusmann, Ralph (2006). Extensible Dependency Grammar: A Modular Grammar Formalism Based On Multigraph Description. PhD. thesis, Saarland University, Germany. find
7. Forsberg, Markus and Aarne Ranta (2004). ‘Functional Morphology’. In Proceedings of ICFP 2004, pages 213–223. ACM Press. find
8. Graff, David and Ke Chen and Junbo Kong and Kazuaki Maeda (2006). Arabic Gigaword Second Edition. LDC catalog number LDC2006T02, ISBN 1-58563-371-2. find
9. Habash, Nizar and Owen Rambow (2005). ‘Arabic Tokenization, Part-of-Speech Tagging and Morphological Disambiguation in One Fell Swoop’. In Proceedings of the 43rd Annual Meeting of the Association for Computational Linguistics ACL 2005, pages 573–580, Ann Arbor. find our card
10. Habash, Nizar and Owen Rambow (2006). ‘MAGEAD: A Morphological Analyzer and Generator for the Arabic Dialects’. In Proceedings of the 21st International Conference on Computational Linguistics and 44th Annual Meeting of the Association for Computational Linguistics, pages 681–688. Association for Computational Linguistics. Sydney, Australia. find our card
11. Hajič, Jan and Otakar Smrž and Tim Buckwalter and Hubert Jin (2005). ‘Feature-Based Tagger of Approximations of Functional Arabic Morphology’. In Proceedings of TLT 2005, pages 53–64, Barcelona, Spain. find
12. Hajič, Jan and Otakar Smrž and Petr Zemánek and Petr Pajas and Jan Šnaidauf and Emanuel Beška and Jakub Kráčmar and Kamila Hassanová (2004). Prague Arabic Dependency Treebank 1.0. LDC catalog number LDC2004T23, ISBN 1-58563-319-4. find
13. Hudak, Paul (1998). ‘Modular Domain Specific Languages and Tools’. In Proceedings of the Fifth International Conference on Software Reuse, pages 134–142. IEEE Computer Society Press. find
14. Lagally, Klaus (1994). Using TeX as a Tool in the Production of a Multi-Lingual Dictionary. Technical Report 1994/15, Fakultät Informatik, Universität Stuttgart. find
15. Lagally, Klaus (2004). ArabTeX: Typesetting Arabic and Hebrew, User Manual Version 4.00. Technical Report 2004/03, Fakultät Informatik, Universität Stuttgart. find
16. Màrquez, Lluís and Dan Klein, eds. (2006). Proceedings of CoNLL-X, the Tenth Conference on Natural Language Learning. ACL, New York. find
17. Smith, Noah A. and David A. Smith and Roy W. Tromble (2005). ‘Context-Based Morphological Disambiguation with Random Fields’. In Proceedings of HLT/EMNLP 2005, pages 475–482, Vancouver, Canada. find
19. Smrž, Otakar and Petr Pajas (2004). ‘MorphoTrees of Arabic and Their Annotation in the TrEd Environment’. In Proceedings of the NEMLAR Conference 2004, pages 38–41, Egypt. find
20. Smrž, Otakar and Petr Zemánek and Jakub Kráčmar and Viktor Bielický (2006). ‘Information Structure with the Prague Arabic Dependency Treebank’. In Proceedings of the Conference on Communication and Information Structure in Spoken Arabic, College Park, Maryland. find
21. Žabokrtský, Zdeněk and Otakar Smrž (2003). ‘Arabic Syntactic Trees: from Constituency to Dependency’. In EACL 2003 Conference Companion, pages 183–186, Budapest, Hungary. find our card

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/aclbulk-2006-smrz-tips-tricks-prague-arabic-dependency/