Slovenian and Spanish

Pending tasks

Testing framework

• Fill out the Pending tests page with some contrastive features of Slovenian and Spanish.
• Set up the testvoc and generation tests.

Slovenian monolingual dictionary

• Proofread pronouns in the Slovenian monolingual dictionary -- They should be complete, and have appropriate subcategorisation (pos, ind, def, qnt, dem, ...)
• Proofread prepositions in the Slovenian monolingual dictionary -- Take note that the same preposition with different forms should be included in a paradigm (e.g. "de" "d'" etc.)
  • Preposition/pronoun contractions should be added as in other pairs, with <j/>
• Proofread conjunctions in the Slovenian dictionary -- They should be split into coordinating and subordinating
• Proofread determiners in the Slovenian dictionary -- They should be complete, and have appropriate subcategorisation (pos, ind, def, qnt, dem, ...)
  • Some determiners are missing (e.g. the possessives moj, naš, vaš, etc.), these need to be added.
• Proofread open categories
  • Verbs are missing aspect markers (perf, imperf) these will need to be added.
  • Verbs are not marked for transitivity.
  • Nouns are not split in moški spol into animate/inanimate. -- They should be
  • Adjectives with superlatives output superlative lemma e.g. najstar -- these should be changed to output non superlative lemma. e.g. najstar → star. See e.g. example for lep, najlep
  • Proper nouns need to be split from common nouns and appropriately subcategorised (top, ant, ...), proper names should be with initial caps.

Disambiguation

• Think of rules to fix frequent disambiguation problems

Slovenian→Spanish bilingual dictionary

• Add closed categories
• Add open categories, remembering to take note of multiple translation possibilities

Slovenian→Spanish transfer

(wait for the sl-it system)

• Move to 3+ levels of transfer
• Investigate how definiteness in Spanish can be inferred from Slovenian.
  • Perhaps consider marking count/mass in bilingual dictionary.
  • Are there any NP structures which are inherently definite ?
    • Superlatives
    • NPs with proper names (e.g. Parlament Slovenije → El Parlamento de Slovenia)

See also

• Pending tests
• Regression tests

Further reading

• Barbara Gawrońska-Werngren (1990) "Translation great problem": on the problem of inserting articles when translating from Russian into Swedish"

• Mel'čuk and Wanner (2008) "Morphological mismatches in machine translation". Machine Translation (§6.1)

• G.R.Martins (1964) "Preliminary report on the insertion of English articles in Russian-English MT output". Mechanical Translation, vol. 8, no.1, August 1964; pp.2-4

• Cornelia Zelinsky-Wibbelt (1992) "Exploiting Linguistic Iconism for Article Machine Translation". University of the Saarlnad, FRG; PROC. OF COLING-92, NANTES. AUG. 23-28, 1992

• Lee Hian-Beng (1999) "Article Selection Using Probabilistic Sense Disambiguation". DSO National Laboratories; 20 Science Park Drive, Singapore 118230; September 1999

• Jocelyn Brewer (1966) "English Article Insertion*". Mechanical Translation and Computational Linguistics, vol.9, nos.3 and 4, Colorado State University, Fort Collins, September and December 1966