Constructing a Lexicon of English Discourse Connectives

What it is about

DiMLex-Eng is a lexicon of 149 English discourse connectives in the XML format of the German DiMLex. Entries were collected from the 100 connectives annotated in PDTB 2.0, the signals of the RST Signalling Corpus (where only about 18% of relations carry a connective or marker), and a list of 230 relational n-grams mined from RST-DT, filtered by a definition: not inflectable, meaning is a two-place relation, both arguments are abstract objects (usually clauses), the expression is fixed (not «for this excellent reason») and not compositional (even though is in; especially when or and therefore are not). Each entry gives orthographic variants, syntactic category (adverb, subordinator/preposition, coordinator, phrase), whether the word has non-connective uses, and the relations it signals with corpus frequencies (e.g. by contrast: Contrast 27 times, split into juxtaposition and opposition). Adding the RST signals raised coverage of RST-SC connective tokens from 87% to 95%.

Ideas

• I1. Move Mova's connective knowledge out of code into one small data file of its own, in the paper's shape: connective, variants, category (mark, cc, advmod, phrase), «has non-connective uses» (since as time vs reason, as, while, so), and the Mova link(s) it can create (cause, next, cond, quote, none) with the condition that selects each. Today the table lives in prag/src/frames.rs and chapter 36 §11.1. Write it from Mova's own seeds and the chapter, not from PDTB counts; the frame reader reads it, and the chapter table is generated from it so the two cannot drift.
• I2. Apply the paper's two exclusion criteria to Mova's cue lists: fixed-expression and non-compositionality. even though / even if get their own entries; intensified forms (just because, only after) are parsed as the base connective plus a modifier. A test lists every multiword connective and checks the parser treats it as one fixed unit.
• I3. Use the «non-connective use» flag as a gate in the reader: for ambiguous connectives (since, as, while, so) the link is created only when the syntactic test for the connective reading passes (a finite clause follows; so not before an adjective), and the decision prints which test fired. Measure false cause links on the frame bench (precision of the cause role).

For Mova

languages/en/36-discourse-relations.md §6 (English connectives and their syntax), §11.1, §11.6 (connective senses not disambiguated). Code: prag/src/frames.rs (connective table → data file), en/seeds/verbal/subordinators-mark.md, dialects/mova/seeds/discourse-adverbs.md. Measures: frame bench cause/next precision; a unit test over the connective list.

Library summary

The Research library's own summary of the paper, in our words — not the paper's text.

Topics: Discourse analysis

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.18653/v1/w18-5042

@inproceedings{das2018constructing,
  title = {{Constructing a Lexicon of English Discourse Connectives}},
  author = {Das, Debopam and Scheffler, Tatjana and Bourgonje, Peter and Stede, Manfred},
  booktitle = {Proceedings of the 19th Annual SIGdial Meeting on Discourse and Dialogue},
  year = {2018},
  doi = {10.18653/v1/w18-5042},
  url = {https://doi.org/10.18653/v1/w18-5042}
}

References

20 references, 4 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Ziheng Lin, Hwee Tou Ng, and Min-Yen Kan. 2014. A PDTB-Styled End-to-End Discourse Parser. Natural Language Engineering, 20(2):151184. find
2. Laura Alonso i Alemany. 2005. Representing discourse for automatic text summarization via shallow NLP techniques. PhD dissertation, Universitat de Barcelona. find
3. William C. Mann and Sandra A. Thompson. 1988. Rhetorical structure theory: Toward a functional theory of text organization. Text, 8(3):243–281. ↗ our card
4. Or Biran and Owen Rambow. 2011. Dentifying justifications in written dialogs by classifying text as argumentative. International Journal of Semantic Computing, 5(4):363–381. find
5. Amalia Mendes, Iria del Rio, Manfred Stede, and Felix Dombek. 2018. A Lexicon of Discourse Markers for Portuguese: LDM-PTs. In Proceedings of the 11h International Conference on Language Resources and Evaluation (LREC), Miyazaki, Japan. find
6. Thomas Meyer, Andrei Popescu-Belis, Sandrine Zufferey, and Bruno Cartoni. 2011. Multilingual Annotation and Disambiguation of Discourse Connectives for Machine Translation. In Proceedings of the SIG-DIAL 2011 Conference, SIGDIAL ’11, pages 194– 203, Stroudsburg, PA, USA. Association for Computational Linguistics. find
7. Peter Bourgonje, Yulia Grishina, and Manfred Stede. 2017. Toward a bilingual lexical database on connectives: Exploiting a German/Italian parallel corpus. In Proceedings of the Fourth Italian Conference on Computational Linguistics, Rome, Italy. find
8. Antonio Briz, Salvador Pons Borderı́a, and José Por-tolés. 2008. Diccionario de partı́culas discursivas del español. find
9. Jir̆ı́ Mı́rovský, Pavlı́na Synková, Magdaléna Rysová, and Lucie Polı́ková. 2017. CzeDLex A Lexicon of Czech Discourse Connectives. In the Prague Bulletin of Mathematical Linguistics, volume 109, pages 61–91. find
10. Lynn Carlson, Daniel Marcu, and Mary Ellen Okurowski. 2003. Building a discourse-tagged corpus in the framework of Rhetorical Structure Theory. In Jan van Kuppevelt and Ronnie Smith, editors, Current Directions in Discourse and Dialogue. Kluwer, Dordrecht. find our card
11. Renate Pasch, Ursula Braue, Eva Breindl, and Herrmann Ulrich Waner. 2003. Handbuch der deutschen Konnektoren. Walter de Gruyter, Berlin/New York. find
12. Bruno Cartoni, Sandrine Zufferey, and Thomas Meyer. 2013. Annotating the meaning of discourse connectives by looking at their translation: The translation-spotting technique. 4:65–86. find
13. Rashmi Prasad, Nikhil Dinesh, Alan Lee, Eleni Milt-sakaki, Livio Robaldo, Aravind Joshi, and Bonnie Webber. 2008. The Penn Discourse Treebank 2.0. In Proceedings of the 6th International Conference on Language Resources and Evaluation (LREC 2008), Marrackech, Morocco. find our card
14. Laurence Danlos, Kateřina Rysová, Magdaléna Rysová, and Manfred Stede. 2018. Primary and secondary discourse connectives: definitions and lexicons. Dialogue and Discourse. To appear. find
15. Charlotte Roze, Danlos Laurence, and Philippe Muller. 2010. LEXCONN: a French Lexicon of Discourse Connectives. In Multidisciplinary Approaches to Discourse - MAD 2010, Moissac, France. find
16. Debopam Das, Maite Taboada, and Paul McFetridge. 2015. RST Signalling Corpus, LDC2015T10. Linguistic Data Consortium. find
17. Tatjana Scheffler and Manfred Stede. 2016. Adding Semantic Relations to a Large-Coverage Connective Lexicon of German. In Proceedings of the Tenth International Conference on Language Resources and Evaluation (LREC), Portoroz̆, Slovenia. find
18. Anna Feltracco, Elisabetta Jezek, Bernardo Magnini, and Manfred Stede. 2016. LICO: A Lexicon of Italian Connectives. In Proceedings of the Second Italian Conference on Computational Linguistic (CLiC-it 2016). find
19. Manfred Stede and Carla Umbach. 1998. DiMLex: A lexicon of discourse markers for text generation and understanding. In Proceedings of the 17th international conference on Computational linguistics - Volume 2. Association for Computational Linguistics. find
20. Hugo Hernault, Hemut Prendinger, David duVerle, and Mitsuru Ishizuka. 2010. HILDA: A Discourse Parser Using Support Vector Machine Classification. Dialogue and Discourse, 1(3):1–33. find our card