Representing Text Meaning for Multilingual Knowledge-Based Machine Translation

What it is about

The Mikrokosmos project defines a text meaning representation (TMR) as its interlingua. A TMR is a frame notation: each clause becomes an instance of an ontology event or property, with slots for time, aspect and polarity, and modifying roles. Pragmatic content sits beside the propositions. Six attitude types (epistemic, deontic, volition, expectation, evaluative, potential) are recorded with an attributed agent, a scope, a time, and a value between 0 and 1 for how positive or certain the attitude is. Domain relations (causal, conjunction, elaboration, alternation) and temporal relations link the frames. The authors built the representation by hand-annotating about 1,300 Japanese business newswires, and used that field work to refine the notation. The paper reports the notation and its experience, with no automatic scores.

Ideas

• I1. Give the pragmatic layer graded values. Each event frame in prag gets a value between 0 and 1 for an attitude such as epistemic or deontic, set by counted cues (modal verbs, negation, reported speech). A small integer scale keeps it deterministic.
• I2. Type the links between frames. The paper's four relation types (causal, conjunction, elaboration, alternation) can be counted from connectives in the Bible verses, such as "therefore" or "if", and stored as typed links in the world store.
• I3. Hand-annotate a small gold set, 50 verses at most, for attitude values, and keep it held out. It serves as the control for the pragmatic layer and must never be used for training.

For Mova

The attitude values and typed links belong in prag (the roles) and in world, with the frames from tools/roundtrip. They also answer open problem 4, function words and boundaries as features, for the speech and clause-boundary types. The hand-annotated set is the only part that needs a human, so it must stay small and separate from the training data.

Topics: Machine translation

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.21236/ada302333

@inproceedings{carlson1994representing,
  title = {{Representing Text Meaning for Multilingual Knowledge-Based Machine Translation}},
  author = {Carlson, Lynn and Cooper, Elizabeth and Dolan, Ronald and Maiorano, Steven},
  booktitle = {Proceedings of the First Conference of the Association for Machine Translation in the Americas},
  year = {1994},
  doi = {10.21236/ada302333},
  url = {https://doi.org/10.21236/ada302333}
}

References

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

1. Brown, R. 1994. FRAMEPAC. Center for Machine Translation. Carnegie Mellon University. CMU-CMT-MEMO. find
2. Carlson, L. and S. Nirenburg. 1990. World Modeling for NLP. Center for Machine Translation, Carnegie Mellon University, Tech Report CMU-CMT-90-121. find
3. Dahlgren, K. 1988. Naive Semantics for Natural Language Understanding. Boston, MA: Kluwer Academic Press. Executive Office of the President. Office of Management and Budget. ↗
4. Standard Industrial Classification Manual, 1987. Springfield, Virginia. find
5. Goodman, K. and S. Nirenburg (eds.). 1991. The KBMT Project: A Case Study in Knowledge-Based Machine Translation. San Mateo, CA: Morgan Kaufmann. find
6. Hirst, G. 1989. Ontological Assumptions in Knowledge Representation. In Proceedings of the First International Conference on Principles of Knowledge Representation and Reasoning, Toronto, Canada, 157-169. find
7. Hovy, E. 1988. Generating Natural Language Under Pragmatic Constraints. Ph.D. Dissertation. Yale University. ↗
8. Hovy, E. and E. Maier. 1994. Parsimonious or Profligate: How Many and Which Discourse Structure Relations? Discourse Processes (to appear). find
9. Lenat, G. and R. Guha. 1990. Building Large Knowledge-Based Systems. Reading, MA: Addison-Wesley Meyer, I., B. Onyshkevych and L. Carlson. 1990. Lexicographic Principles and Design for Knowledge-Based Machine Translation. Technical Report (CMU-CMT-90-118), Center for Machine Translation. Carnegie Mellon University, Pittsburgh, PA. find
10. Miller, G. 1990. Wordnet: An on-line lexical database. International Journal of Lexicography 3(4). find
11. Nirenburg, S., J. Carbonell, M. Tomita and K. Goodman. 1992. Machine Translation: A Knowledge-Based Approach. San Mateo, CA: Morgan Kaufmann. find
12. Nirenburg, S. and C. Defrise. 1993. Lexical and Conceptual Structure for Knowledge-Based Machine Translation. In J. Pustejovsky (ed.), Semantics and the Lexicon. Dordrecht: Kluwer. ↗
13. Nirenburg, S. and L. Levin. 1991. Syntax-Driven and Ontology-Driven Lexical Semantics. In Lexical Semantics and Knowledge Representation: Proceedings of a Workshop Sponsored by the Special Interest Group on the Lexicon of the Association for Computational Linguistics, Berkeley, CA, 9-19. find
14. Nirenburg, S. and J. Pustejovsky. 1988. Processing Aspectual Semantics, Proceedings of the Tenth Annual Meeting of the Cognitive Science Society, Montreal, Canada. find
15. Nyberg, E. 1988. The FRAMEKIT User's Guide, Version 2.0. Center for Machine Translation, Carnegie Mellon University. CMU-CMT-MEMO. find
16. Onyshkevych, B. and S. Nirenburg. 1991. Lexicon, Ontology and Text Meaning. In Lexical Semantics and Knowledge Representation: Proceedings of a Workshop Sponsored by the Special Interest Group on the Lexicon of the Association for Computational Linguistics, Berkeley, CA, 238-249. find
17. Tipster Text Program. 1993. Phase I: Proceedings of a Workshop held at Fredericksburg, Virginia, Morgan Kaufmann. find
18. Tokyo Keizai Inc. 1990. Japan Company Handbook. Tokyo, Japan. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/carlson-1994-representing-text-meaning-multilingual-knowledge/