The Use of Parallel and Comparable Data for Analysis of Abstract Anaphora in German and English

What it is about

Abstract anaphora (a pronoun or a demonstrative noun phrase that refers to an event, fact or proposition) is a training target for anaphora resolution. The question is whether translated English text, here English turns that were translated from German in the Europarl corpus, is "authentic" enough to train on, or whether only original English should be used. The authors compare four subcorpora (original and translated, in each language) with the same parliamentary register. They measure sentence length, nouns per token, definite and demonstrative noun chunks, and the use of 211 English (452 German) "label nouns" such as fact, issue, proposal, which stand in for abstract anaphora. They use Cohen's d rather than significance tests, because the samples are large. The result: English translations differ from English originals with d = -0.15 for sentence length (small), while the German-English pairs differ with d = -0.31 for original against original and -0.44 for original German against translated English, which the authors read as explicitation. German translations show larger shifts in noun and definite-NC use. English translations are judged usable for training, German less clearly.

Ideas

• I1. A translationese check before training: compare the sentence length, the nouns-per-token rate and the demonstrative-NP rate of any translated English corpus against an original English baseline, with Cohen's d as the gate (small effect, d below 0.3, is accepted). Log the d values with the run.
• I2. A closed lexicon of label nouns as an enum in en/src/dict.rs (fact, issue, proposal, task, reply, criticism, and so on), used by a rule: a demonstrative plus a label noun ("this task") gets an abstract-antecedent flag, and the antecedent search looks for a clause or a sentence rather than a noun.
• I3. Negative control: a demonstrative plus a concrete noun ("this chair") must not get the abstract flag. Count the false flags on a set of concrete demonstratives.
• I4. Explicitation as a feature: count how often an English text uses a full definite NP where a pronoun would do, and check whether the count rises in translated text; use the rate as a flag, not a score.

For Mova

The chapters are 35. Reference and coreference (abstract anaphora, demonstratives and reference to events), 12. Pronouns, determiners and numerals (the this/that determiners) and 37. Register and variation (register and the effect of translation). The code is coref/src/mention.rs (demonstrative mentions), coref/src/sieve.rs (the abstract antecedent rule) and en/src/dict.rs (the label-noun class). The measures: Cohen's d between Mova's own original and translated English texts on the four features, with the translated-text gate; precision of the abstract-anaphor flag on a hand-labelled set, with the concrete-demonstrative negative control. The paper's Europarl data are evaluation-only, so Mova should build its own original-English baseline.

Topics: Coreference resolution

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.63317/2n46634asxy9

@inproceedings{dipper2012use,
  title = {{The Use of Parallel and Comparable Data for Analysis of Abstract Anaphora in German and English}},
  author = {Dipper, Stefanie and Seiss, Melanie and Zinsmeister, Heike},
  booktitle = {Proceedings of the Eighth International Conference on Language Resources and Evaluation (LREC'12)},
  year = {2012},
  doi = {10.63317/2n46634asxy9},
  url = {https://doi.org/10.63317/2n46634asxy9}
}

References

37 references, 2 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Lars Ahrenberg, Jörg Tiedemann, and Martin Volk, editors. 2010. Proceedings of the Workshop on Annotation and Exploitation of Parallel Corpora, volume 10 of NEALT proceedings series, Tartu, Estonia. find
2. Nicholas Asher. 1993. Reference to Abstract Objects in Discourse. Kluwer Academic Publishers, Boston MA. find
3. Mona Baker. 1993. Corpus linguistics and translation studies: implications and applications. In Mona Baker, Gill Francis, and Elena Tognini-Bonelli, editors, Text and Technology: In Honour of John Sinclair, pages 233–250. John Benjamins, Amsterdam/Philadelphia. Freq orig 255 (226) 642 (569) 168 (149) 717 (635) 281 (249) Freq trans 1803 (2106) 2782 (3250) 666 (778) 2096 (2449) 807 (943) Increase 7.1 4.3 4.0 2.9 2.9 Decrease 2.8 2.6 2.0 2.0 1.9 624 395 951 432 304 224 152 469 218 161 (262) (178) (548) (255) (188) (553) (350) (843) (383) (269) Freq orig 201 (133) 132 (87) 713 (471) 413 (273) 1008 (666) Freq trans 921 (795) 424 (366) 1871 (1615) 837 (722) 2021 (1744) Increase 4.6 3.2 2.6 2.0 2.0 Decrease 2.5 2.5 2.3 2.2 2.1 684 979 958 863 427 269 389 409 397 199 (452) (647) (633) (570) (282) (232) (336) (353) (343) (172) find
4. Viktor Becher. 2011. Explicitation and implicitation in translation. A corpus-based study of English–German and German–English translations of business texts. Ph.D. thesis, Universität Hamburg. find
5. Luisa Bentivogli and Emanuele Pianta. 2005. Exploiting parallel texts in the creation of multilingual semantically annotated resources: the MultiSemCor corpus. Natural Language Engineering, 11(3). find
6. Shoshana Blum-Kulka. 1986. Shifts of cohesion and coherence in translation. In Juliane House and Shoshana Blum-Kulka, editors, Interlingual and intercultural communication, pages 17–35. Tübingen: Gunter Narr. find
7. Sabine Brants, Stefanie Dipper, Peter Eisenberg, Silvia Hansen, Esther König, Wolfgang Lezius, Christian Rohrer, George Smith, and Hans Uszkoreit. 2004. TIGER: Linguistic Interpretation of a German Corpus. Journal of Language and Computation, 2(4):597–620. Special Issue. find
8. Donna K. Byron. 2002. Resolving pronominal reference to abstract entities. In Proceedings of the ACL-02 conference, pages 80–87. find
9. Bruno Cartoni, Sandrine Zufferey, Thomas Meyer, and Andrei Popescu-Belis. 2011. How comparable are parallel corpora? Measuring the distribution of general vocabulary and connectives. In Proceedings of 4th Workshop on Building and Using Comparable Corpora, at ACL-HLT 2011, pages 78–86. find
10. José Castaño, Jason Zhang, and James Pustejovsky. 2002. Anaphora resolution in biomedical literature. In Proceedings of the International Symposium on Reference Resolution for NLP. find
11. Jacob Cohen. 1988. Statistical Power Analysis for the Behavioral Sciences. Lawrence Erlbaum Associates, Hillsdale, New Jersey, 2nd edition. find
12. Oliver Čulo, Silvia Hansen-Schirra, Stella Neumann, and Mihaela Vela. 2008. Empirical studies on language contrast using the English-German comparable and parallel CroCo corpus. In Proceedings of LREC 2008. find
13. Stefanie Dipper and Heike Zinsmeister. 2010. Towards a standard for annotating abstract anaphora. In Proceedings of the LREC-Workshop Language Resource and Language Technology Standards — state of the art, emerging needs, and future developments. Stefanie Dipper and Heike Zinsmeister. to appear. Annotating abstract anaphora. find
14. Language Resources and Evaluation, Online First Sep 2011. find
15. Stefanie Dipper, Christine Rieger, Melanie Seiss, and Heike Zinsmeister. 2011. Abstract anaphors in German and English. In Iris Hendrickx, Sobha Lalitha Devi, António Branco, and Ruslan Mitkov, editors, Anaphora Processing and Applications: 8th Discourse Anaphora and Anaphor Resolution Colloquium, DAARC 2011. Revised selected papers, pages 96–107. Springer. find
16. Bonnie J. Dorr. 1994. Machine translation divergences: A formal description and proposed solution. Computational Linguistics, 20(4):597–633. find our card
17. Gill Francis. 1994. Labelling discourse: an aspect of nominal group lexical cohesion. In Malcolm Coulthard, editor, Advances in Written Text Analysis, pages 83–101. London: Routledge. find
18. Kari Fraurud. 1992. Situation reference: What does ’it’ refer to? GAP Working Paper No 24, Fachbereich Informatik, Universität Hamburg. find
19. Stephan Th. Gries. 2005. Null-hypothesis significance testing of word frequencies: A follow-up on Kilgarriff. Corpus Linguistics and Linguistic Theory, 1:277–294. find
20. Hans van Halteren. 2008. Source language markers in EU-ROPARL translations. In Proceedings of the 22nd International Conference on Computational Linguistics COL-ING 08, pages 937–944. find
21. Andreas Kaster, Stefan Siersdorfer, and Gerhard Weikum. 2005. Combining text and linguistic document representations for authorship attribution. In SIGIR Workshop: Stylistic Analysis of Text for Information Access (STYLE). find
22. Kinga Klaudy. 2008. Explicitation. In Mona Baker and Gabriela Saldanha, editors, Routledge Encyclopedia of Translation Studies, pages 104–108. London and New York: Routledge, 2nd edition. find
23. Philipp Koehn. 2005. Europarl: A parallel corpus for statistical machine translation. In Proceedings of MT Summit. find
24. Iørn Korzen and Morten Gylling. 2011. What can contrastive linguistics tell us about translating discourse structure? In Book of Abstracts of the GSCL Workshop on Contrastive Linguistics, Translation Studies, Machine Translation — what can we learn from each other? find
25. Mitchell Marcus, Beatrice Santorini, and Mary Ann Marcinkiewicz. 1993. Building a large annotated corpus of English: The Penn Treebank. Computational Linguistics, 19(2):313–330. find our card
26. Costanza Navarretta and Sussi Olsen. 2008. Annotating abstract pronominal anaphora in the DAD project. In Proceedings of LREC-08. find
27. Costanza Navarretta. 2008. Pronominal types and abstract reference in the Danish and Italian DAD corpora. In Proceedings of the Second Workshop on Anaphora Resolution, pages 63–71. find
28. Sameer Pradhan, Lance Ramshaw, Ralph Weischedel, Jessica MacBride, and Linnea Micciulla. 2007. Unrestricted coreference: Identifying entities and events in OntoNotes. In Proceedings of the IEEE-ICSC. find
29. Marta Recasens. 2008. Discourse deixis and coreference: Evidence from AnCora. In Proceedings of the Second Workshop on Anaphora Resolution, pages 73–82. find
30. Beatrice Santorini. 1990. Part-of-speech tagging guidelines for the Penn Treebank Project. Technical Report MS-CIS-90-47, Department of Computer and Information Science, University of Pennsylvania. find
31. Anne Schiller, Simone Teufel, Christine Stöckert, and Christine Thielen. 1999. Guidelines für das Tagging deutscher Textcorpora mit STTS (kleines und großes Tagset). Technical report, University of Stuttgart and University of Tübingen. find
32. Helmut Schmid. 1994. Probabilistic part-of-speech tagging using decision tree. In Proceedings of International Conference on New Methods in Language Processing. find
33. Kiril Simov, Petya Osenova, Radovan Garabik, and Jörg Tiedemann, editors. 2011. Proceedings of The Second-Workshop on Annotation and Exploitation of Parallel Corpora, Hissar, Bulgaria. find
34. Elke Teich. 2003. Cross-linguistic Variation in System and Text: A Methodology for the Investigation of Translations and Comparable Texts. Mouton de Gruyter, Berlin. find
35. Renata Vieira, Susanne Salmon-Alt, and Caroline Gasperin. 2002. Coreference and anaphoric relations of demonstrative noun phrases in a multilingual corpus. In Proceedings of DAARC-2002. find
36. Jean-Paul Vinay and Jean Darbelnet. 1995. Comparative stylistics of French and English: A methodology for translation. John Benjamins, Amsterdam/Philadelphia. find
37. Ralph Weischedel et al. 2010. OntoNotes Release 4.0, with OntoNotes DB Tool v. 0.999 beta. Technical report, Raytheon BBN Technologies et al. Heike Zinsmeister, Stefanie Dipper, and Melanie Seiss. submitted. Abstract pronominal anaphors and label nouns in German and English: selected case studies and quantitative investigations. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/dipper-2012-use-parallel-comparable-data-analysis/