Translation by Structural Correspondences

What it is about

The paper proposes machine translation without a transfer program. The source and target sentences are described by LFG structures at several levels (constituent, functional, semantic), and the translation is a set of correspondence functions between those structures. The functions are written as equations, so the mapping can depend on information from several levels at once, and both grammars and lexicons stay modular. The authors show the method with English, German and French examples where the source and target analyses differ in structure. The paper is a sketch of the approach; it has no evaluation.

Ideas

• I1. Store the author delta as a set of partial correspondences between the features of meta tokens, not as a procedure. Each override names a source feature, a target feature and a condition, and the encoder solves the set. Conflicts between overrides are then counted rather than hidden by order.
• I2. Use two levels at once: keep the UD relation and the frame role on each meta token, and learn the correspondence from role to target relation (for example, subject to subject) from aligned verses. Count how often one role maps to one relation in each author pair.
• I3. Measure which residual class the missing correspondences cause: if a verse fails only because a role has no mapping, the error belongs to the mapping, not to the reader or the encoder. Report the count per class each run, twice, for determinism.

For Mova

The code is tools/translate (meta.rs, profile.rs) with the roles of prag. The paper is not a method Mova should copy, since its equations are hand-written, but the idea of several simultaneous levels with partial mappings is the form the open problems 1 and 3 ask for.

Topics: Machine translation

Cite

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

@inproceedings{kaplan1989translation,
  title = {{Translation by Structural Correspondences}},
  author = {Kaplan, Ronald M. and Netter, Klaus and Wedekind, Jurgen and Zaenen, Annie},
  booktitle = {Fourth Conference of the European Chapter of the Association for Computational Linguistics},
  year = {1989},
  url = {https://aclanthology.org/E89-1037/}
}

References

17 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Arnold, Douglas and Louis des Tombe (1987) Basic theory and methodology in Eurotra. In S. Nirenburg (ed.), Machine translation: Theoretical and methodological issues. Cambridge: Cambridge University Press, 114-135. find
2. Fenstad, Jens Erik, Per-Kristian Halvorsen, Tore Langholm, and Johan van Benthem (1987) Situations, Language and Logic. Dordrecht: D. Reidel. find
3. Halvorsen, Per-Kristian (1983) Semantics for lexical-functional grammars. Linguistic Inquiry, 14 (3), 567-613. find
4. Halvorsen, Per-Kristian (1988) Situation semantics and semantic interpreation in constraint-base grammars. Proceedings of the International Conference on Fifth Generation Computer Systems, Tokyo, Japan, 471-478. find
5. Halvorsen, Per-Kristian and Ronald Kaplan (1988) Projections and semantic description. Proceedings of the International Conference on Fifth Generation Computer Systems, Tokyo, Japan, 1116-1122. find
6. Isabelle, Pierre and Elliott Macklovitch (1986) Transfer and M T modularity. Proceedings of Coling I986, Bonn, 115-117. - 281 find
7. Kaplan, Ronald (1987) Three seductions of Computational Pyscholinguistics. In Peter Whitelock et al. (eds.), Linguistic Theory and Computer Applications. Academic Press, London, 149-188. find
8. Kaplan, Ronald (1988) Correspondences and their inverses. Paper presented at the Syntax and Semantics Workshop, April, Titisee, FRG. find
9. Kaplan, Ronald and Joan Bresnan (1982) Lexical Functional Grammar: a formal system for Grammatical representation. In Joan Bresnan (ed.), The Mental Representation of Grammatical Relations. MIT Press, Cambridge, Mass, 173- 281. find
10. Kaplan, Ronald and John Maxwell (1988a). Constituent coordination in Lexical-Functional Grammar. Proceedings of COLING 88, Budapest, 303-305. find
11. Kaplan, Ronald and John Maxwell (1988b). An algorithm for Functional Uncertainty. Proceedings of COLING 88, Budapest, 297-302. find
12. Kay, Martin (1984) Functional Unification Grammar: A formalism for Machine Translation. Proceedings of Coling 1984, Stanford University, 75-78. find
13. Kudo, Ikuo and Hirosato Nomura (1986) Lexical-Functional Transfer: A Transfer Framework in a Machine Translation System based on LFG. Proceedings of Coling 1986, Bonn, 112-114. find
14. Momma, Stefan and Jochen D6rre (1987) Generation from f-structures. In Ewan Klein and Johan van Benthem (eds.) Categories, Polymorphism and Unification, Edinburgh, Amsterdam, 147-168. find
15. Netter, Klaus and Jiirgen Wedekind (1986) An LFG-based Approach to Machine Translation. Proceedings of IAI-MT 86, SaarbrQcken, 197-209. find
17. Reyle, Uwe (1988) Compositional semantics for LFG. In Uwe Reyle and Christian Rohrer (eds.), Natural language parsing and linguistic theories. Dordrecht: D. Reidel, 448-479. find
18. Wedekind, JQrgen (1988) Generation as Structure Driven Derivation. Proceedings of COLING 88, Budapest, 732-737. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/kaplan-1989-translation-structural-correspondences/