OpenLogos Semantico-Syntactic Knowledge-Rich Bilingual Dictionaries
Summary
The authors release three bilingual dictionaries from the open-source machine translation system OpenLogos: English-German, English-French and English-Italian. The Logos system was created by Bernard Scott in the early 1970s. Its dictionaries were used for several decades in a commercial product, so they were checked by both the developers and the customers. Ordinary dictionaries usually give only part of speech, gender and number. In these resources, each entry also carries semantico-syntactic annotation in the SAL language. SAL is a hierarchical ontology in which the meaning and the syntactic behavior of a word are encoded at the same level. It helps distinguish word senses already at the lexical level. In addition, the dictionaries contain morphological paradigms for both languages, as well as markers for homographs, the head word in multiword units, the auxiliary verb, causativity, reflexivity and aspect. The paper explains the principles of SAL and illustrates the XML structure of the entries with examples. The resources are freely available for research through METANET. The authors consider them useful for other translation systems, for other NLP tasks and for low-resource languages.
Key points
- SAL comprises more than 1 000 categories covering all parts of speech. They are grouped on three levels: superset, set and subset. In the dictionary the categories are stored as numeric codes, while in the paper they are given as mnemonics for convenience (for example, ANrept, INMO, AVquan).
- How SAL distinguishes senses: table has two codes, COsurf (a concrete object with a surface) and INdata (recorded data). English crocodile is translated into German as Ziehbank when it refers to a mechanism and as Krokodil when it refers to the reptile.
- The idea behind SAL is that a word's semantics influences the syntax around it. For example, adjectives of the easy type and of the eager type behave differently in a sentence, so they receive different codes.
- Each word is linked to a morphological paradigm (PAT) whose model is a typical word: walk, book, parler, Hand. This makes it possible to map word forms between the source and target languages, which helps with agreement.
- The dictionaries are provided in XML format. The authors also plan to release English-Portuguese, English-Spanish and German-English dictionaries.
The quantitative section of the paper (Section 5) is missing from the provided text, so I do not give the exact number of entries in the dictionaries.
Cite
@inproceedings{barreiro2014openlogos,
title = {{OpenLogos Semantico-Syntactic Knowledge-Rich Bilingual Dictionaries}},
author = {Barreiro, Anabela and Batista, Fernando and Ribeiro, Ricardo and Moniz, Helena and Trancoso, Isabel},
booktitle = {Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC'14)},
year = {2014},
doi = {10.63317/3qinsvcfm6xd},
url = {https://doi.org/10.63317/3qinsvcfm6xd}
}References
- B. Vandecapelle A. Raw and F. Van Eynde. 1988. Eurotra: an overview. Interface. Journal of Applied Linguistics, 1(3):5–32. find
- M-H. Antoni-Lay, G. Francopoulo, and L. Zaysser. 1994. A generic model for reusable lexicons: The GENELEX project. Literary and Linguistic Computing, 9(1):47–54. find
- Collin F. Baker, Charles J. Fillmore, and John B. Lowe. 1998. The berkeley framenet project. In Proceedings of the 36th Annual Meeting of the Association for Computational Linguistics and 17th International Conference on Computational Linguistics - Volume 1, ACL ’98, pages 86–90, Stroudsburg, PA, US… find our card
- Anabela Barreiro, Bernard Scott, Walter Kasper, and Bernd Kiefer. 2011. Openlogos rule-based machine translation: Philosophy, model, resources and customization. Machine Translation, 25(2):107–126. find
- Nicoletta Calzolari, Alessandro Lenci, Francesca Bertagna, and Antonio Zampolli. 2002. Broadening the scope of the eagles/isle lexical standardization initiative. In Proceedings of the 3rd Workshop on Asian Language Resources and International Standardization - Volume 12, COLING ’02, pages 1–8, Stro… find
- Christiane Fellbaum, editor. 1998. WordNet: An Electronic Lexical Database. MIT Press, Cambridge, MA. find
- Rebecca Green, Bonnie J. Dorr, and Philip Resnik. 2004. Inducing frame semantic verb classes from wordnet and ldoce. In Proceedings of the 42nd Meeting of the Association for Computational Linguistics (ACL’04), Main Volume, pages 375–382, Barcelona, Spain, July. find
- Alessandro Lenci, Nuria Bel, Federica Busa, Nicoletta Calzolari, Elisabetta Gola, Monica Monachini, Antoine Ogonowski, Ivonne Peters, Wim Peters, Nilda Ruimy, Marta Villegas, and Antonio Zampolli. 2000. Simple: A general framework for the development of multilingual lexicons. International Journal o… find our card
- George A. Miller. 1995. Wordnet: A lexical database for english. Communications of the ACM, 38(11):39–41. find our card
- Eugenio Picchi, Carol Peters, and Nicoletta Calzolari. 1988. Implementing a Bilingual Lexical Database System. In T. Magay and J. Zigány, editors, Proceedings of the 3rd EURALEX International Congress, pages 317– 329, Budapest, Hungary, sep. Akadémiai Kiadó. find
- Paul Procter, editor. 1978. Longman Dictionary of Contemporary English. Longman Group Ltd., Essex, UK. find
- Bernard (Bud) Scott. 2003. The logos model: An historical perspective. Machine Translation, 18(1):1–72. find
- Marta Villegas, Núria Bel, Alessandro Lenci, Nicoletta Calzolari, Nilda Ruimy, Antonio Zampolli, Teresa Sadurnı́, and Joan Soler. 2000. Multilingual linguistic resources: From monolingual lexicons to bilingual interrelated lexicons. In LREC. European Language Resources Association. find
- P. Vossen. 1998. EuroWordNet: A Multilingual Database with Lexical Semantic Networks. Kluwer Academic Publishers. find
This page in Markdown → https://md.research.elamur.ai/free/en/papers/barreiro-2014-openlogos-semantico-syntactic-knowledge-rich/