---
title: "Towards Cross-Domain PDTB-Style Discourse — Stepanov 2014"
description: "This paper asks whether a discourse parser trained on the news-wire Penn Discourse Treebank (PDTB) can locate argument spans of explicit discourse…"
authors: ["Evgeny Stepanov", "Giuseppe Riccardi"]
year: 2014
venue: "Proceedings of the 5th International Workshop on Health Text Mining and Information Analysis (Louhi)"
source: "https://aclanthology.org/W14-1105/"
doi: "10.3115/v1/w14-1105"
licence: "CC BY-NC-SA 3.0 (ACL Anthology, before 2016)"
licence_url: "https://creativecommons.org/licenses/by-nc-sa/3.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/papers/stepanov-2014-cross-domain-pdtb-style-discourse/"
---
[Elamur Research](https://md.research.elamur.ai/) · [this page in HTML](https://research.elamur.ai/free/papers/stepanov-2014-cross-domain-pdtb-style-discourse/)

[home](https://md.research.elamur.ai/) / [papers](https://md.research.elamur.ai/free/papers/) / [2014](https://md.research.elamur.ai/free/papers/2014/page/2/) / Towards Cross-Domain PDTB-Style Discourse Parsing

# Towards Cross-Domain PDTB-Style Discourse Parsing

**Authors:** [Evgeny Stepanov](https://md.research.elamur.ai/free/authors/evgeny-stepanov/), [Giuseppe Riccardi](https://md.research.elamur.ai/free/authors/giuseppe-riccardi/) · **Year:** 2014 · **Venue:** Proceedings of the 5th International Workshop on Health Text Mining and Information Analysis (Louhi) · **Licence:** [CC BY-NC-SA 3.0 (ACL Anthology, before 2016)](https://creativecommons.org/licenses/by-nc-sa/3.0/)

**Links:** [Publisher's page](https://aclanthology.org/W14-1105/) · [PDF](https://aclanthology.org/W14-1105.pdf) · [DOI 10.3115/v1/w14-1105](https://doi.org/10.3115/v1/w14-1105) · [OpenAlex](https://openalex.org/W2251828655)

**Citations:** 9 citations (openalex-snapshot · 2026-10-14) · [6](#cited-by) in our library, of the 140673 papers whose bibliographies we read

## Summary
This paper asks whether a discourse parser trained on the news-wire Penn Discourse Treebank (PDTB) can locate argument spans of explicit discourse relations in biomedical text from BioDRB. Earlier studies found that connective detection and sense classification transfer poorly between these domains. Argument span extraction had not been tested across domains. The authors take a PDTB-optimized pipeline and apply it to BioDRB, assuming the connective is already identified. A BoosTexter classifier first decides whether Arg1 lies in the same sentence as the connective or in a previous one. CRF sequence labelers then mark each argument's tokens. BioDRB results come from 12-fold cross-validation.

Span extraction turns out to transfer well. However, the PDTB-optimized features rely on connective senses, and sense classification itself generalizes badly. The authors therefore try feature-level adaptation, replacing senses with plain connective labels or using mainly syntactic features. These variants perform about as well as the baseline while removing the dependence on sense prediction. Models trained in-domain on BioDRB still score higher. The study suggests that biomedical discourse parsing can reuse news-trained span models, and that the main remaining obstacles are connective detection and biomedical-specific preprocessing.

## Key points
- **Corpus comparison:**
  - PDTB contains 18,459 explicit relations with 100 connective types.
  - BioDRB covers 24 GENIA articles, with 2,636 relations and 123 connective types.
  - Only about 56% of BioDRB connective types also occur in PDTB.
  - Subordinators, a class that PDTB does not annotate, account for 33% of BioDRB connective types.
- **Argument position classification:** the BoosTexter model reaches F1 = 98.12 on PDTB sections 23–24 with gold preprocessing and 97.81 with Stanford Parser preprocessing.
- **CRF span extraction:**
  - Four CRF models handle the cases SS Arg1/Arg2 and PS Arg1/Arg2.
  - The FS case is dropped because PDTB has only 8 such instances.
  - Sentence selection for the PS case rests on two PDTB patterns: Arg2 sits fully in the connective's sentence in 98.5% of cases, and Arg1 sits fully in the immediately preceding sentence in 71.7%.
- **Earlier cross-domain results:** previously reported cross-domain scores were F1 = 0.55 for connective detection and F1 = 0.57 for sense classification.
- **Adaptation results:**
  - The "No Relation Sense" model for Arg1 slightly outperforms the baseline, though the difference is not significant.
  - A syntactic model with tokens and 2-grams added performs on par with the baseline.
  - Future work will test tools tuned for the biomedical domain.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

Fields: [Linguistics & languages](https://md.research.elamur.ai/free/shelves/linguistics/)

Topics: [Discourse analysis](https://md.research.elamur.ai/free/topics/discourse-analysis/)

## Cited by

Cited by

6 papers of our library cite it (of the 140673 whose bibliographies we read); 3 of them have a page here, 3 not yet.

- [Let's be explicit about that: Distant supervision for implicit discourse relation classification via connective prediction](https://md.research.elamur.ai/free/papers/kurfal-2021-let-s-be-explicit-about/)
- [The UniTN Discourse Parser in CoNLL 2015 Shared Task: Token-level Sequence Labeling with Argument-specific Models](https://md.research.elamur.ai/free/papers/stepanov-2015-unitn-discourse-parser-conll-2015/)
- [UniTN End-to-End Discourse Parser for CoNLL 2016 Shared Task](https://md.research.elamur.ai/free/papers/stepanov-2016-unitn-end-end-discourse-parser/)

## Cite

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

```bibtex
@inproceedings{stepanov2014cross,
  title = {{Towards Cross-Domain PDTB-Style Discourse Parsing}},
  author = {Stepanov, Evgeny and Riccardi, Giuseppe},
  booktitle = {Proceedings of the 5th International Workshop on Health Text Mining and Information Analysis (Louhi)},
  year = {2014},
  doi = {10.3115/v1/w14-1105},
  url = {https://doi.org/10.3115/v1/w14-1105}
}
```

- [← previous The Stanford CoreNLP Natural Language Processing Toolkit](https://md.research.elamur.ai/free/papers/manning-2014-stanford-corenlp-natural-language-processing/)
- [next → Unifying Visual-Semantic Embeddings with Multimodal Neural Language Models](https://md.research.elamur.ai/free/papers/kiros-2014-unifying-visual-semantic-embeddings-multimodal/)
- [↑ back to papers of 2014](https://md.research.elamur.ai/free/papers/2014/page/2/)

## References

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

1. Cornelia Brunner and Thomas Wirth. 2006. Btk expression is controlled by oct and bob. 1/obf. 1. Nucleic acids research, 34(6):1807–1815. [find](https://scholar.google.com/scholar?q=Cornelia+Brunner+and+Thomas+Wirth%2E+2006%2E+Btk+expression+is+controlled+by+oct+and+bob%2E+1%2Fobf%2E+1%2E+Nucleic+acids+research%2C+34%286%29%3A1807%E2%80%931815%2E)
2. Sabine Buchholz. 2000. Readme for perl script chunklink.pl. [find](https://scholar.google.com/scholar?q=Sabine+Buchholz%2E+2000%2E+Readme+for+perl+script+chunklink%2Epl%2E)
3. Syeed Ibn Faiz and Robert E Mercer. 2013. Identifying explicit discourse connectives in text. In Advances in Artificial Intelligence, pages 64–76. Springer. [find](https://scholar.google.com/scholar?q=Syeed+Ibn+Faiz+and+Robert+E+Mercer%2E+2013%2E+Identifying+explicit+discourse+connectives+in+text%2E+In+Advances+in+Artificial+Intelligence%2C+pages+64%E2%80%9376%2E+Springer%2E)
4. Sucheta Ghosh, Richard Johansson, Giuseppe Riccardi, and Sara Tonelli. 2011. Shallow discourse parsing with conditional random fields. In Proceedings of the 5th International Joint Conference on Natural Language Processing (IJCNLP 2011). [find](https://scholar.google.com/scholar?q=Sucheta+Ghosh%2C+Richard+Johansson%2C+Giuseppe+Riccardi%2C+and+Sara+Tonelli%2E+2011%2E+Shallow+discourse+parsing+with+conditional+random+fields%2E+In+Proceedings+of+the+5th+International+Joint+Conference+on+Natur) [our card](https://md.research.elamur.ai/free/papers/ghosh-2011-shallow-discourse-parsing-conditional-random/)
5. J-D Kim, Tomoko Ohta, Yuka Tateisi, and Junichi Tsujii. 2003. Genia corpusa semantically annotated corpus for bio-textmining. Bioinformatics, 19(suppl 1):i180–i182. [find](https://scholar.google.com/scholar?q=J%2DD+Kim%2C+Tomoko+Ohta%2C+Yuka+Tateisi%2C+and+Junichi+Tsujii%2E+2003%2E+Genia+corpusa+semantically+annotated+corpus+for+bio%2Dtextmining%2E+Bioinformatics%2C+19%28suppl+1%29%3Ai180%E2%80%93i182%2E)
6. Dan Klein and Christopher D. Manning. 2003. Fast exact inference with a factored model for natural language parsing. In Advances in Neural Information Processing Systems 15 (NIPS 2002). [find](https://scholar.google.com/scholar?q=Dan+Klein+and+Christopher+D%2E+Manning%2E+2003%2E+Fast+exact+inference+with+a+factored+model+for+natural+language+parsing%2E+In+Advances+in+Neural+Information+Processing+Systems+15+%28NIPS+2002%29%2E)
7. John Lafferty, Andrew McCallum, and Fernando C. N. Pereira. 2001. Conditional random fields: Probabilistic models for segmenting and labeling sequence data. In Proceedings of 18th International Conference on Machine Learning. [find](https://scholar.google.com/scholar?q=John+Lafferty%2C+Andrew+McCallum%2C+and+Fernando+C%2E+N%2E+Pereira%2E+2001%2E+Conditional+random+fields%3A+Probabilistic+models+for+segmenting+and+labeling+sequence+data%2E+In+Proceedings+of+18th+International+Confer)
8. Ziheng Lin, Hwee Tou Ng, and Min-Yen Kan. 2012. A pdtb-styled end-to-end discourse parser. Natural Language Engineering, 1:1 – 35. [find](https://scholar.google.com/scholar?q=Ziheng+Lin%2C+Hwee+Tou+Ng%2C+and+Min%2DYen+Kan%2E+2012%2E+A+pdtb%2Dstyled+end%2Dto%2Dend+discourse+parser%2E+Natural+Language+Engineering%2C+1%3A1+%E2%80%93+35%2E)
9. Rashmi Prasad, Nikhil Dinesh, Alan Lee, Eleni Miltsakaki, 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). [find](https://scholar.google.com/scholar?q=Rashmi+Prasad%2C+Nikhil+Dinesh%2C+Alan+Lee%2C+Eleni+Miltsakaki%2C+Livio+Robaldo%2C+Aravind+Joshi%2C+and+Bonnie+Webber%2E+2008%2E+The+penn+discourse+treebank+2%2E0%2E+In+Proceedings+of+the+6th+International+Conference+on+) [our card](https://md.research.elamur.ai/free/papers/disc-2008-prasad-penn-discourse-treebank-2-0/)
10. Rashmi Prasad, Susan McRoy, Nadya Frid, Aravind Joshi, and Hong Yu. 2011. The biomedical discourse relation bank. BMC Bioinformatics, 12(1):188. [find](https://scholar.google.com/scholar?q=Rashmi+Prasad%2C+Susan+McRoy%2C+Nadya+Frid%2C+Aravind+Joshi%2C+and+Hong+Yu%2E+2011%2E+The+biomedical+discourse+relation+bank%2E+BMC+Bioinformatics%2C+12%281%29%3A188%2E)
11. Balaji Polepalli Ramesh and Hong Yu. 2010. Identifying discourse connectives in biomedical text. AMIA Annual Symposium Proceedings, 2010:657. [find](https://scholar.google.com/scholar?q=Balaji+Polepalli+Ramesh+and+Hong+Yu%2E+2010%2E+Identifying+discourse+connectives+in+biomedical+text%2E+AMIA+Annual+Symposium+Proceedings%2C+2010%3A657%2E)
12. Balaji Polepalli Ramesh, Rashmi Prasad, Tim Miller, Brian Harrington, and Hong Yu. 2012. Automatic discourse connective detection in biomedical text. Journal of the American Medical Informatics Association, 19(5):800–808. [find](https://scholar.google.com/scholar?q=Balaji+Polepalli+Ramesh%2C+Rashmi+Prasad%2C+Tim+Miller%2C+Brian+Harrington%2C+and+Hong+Yu%2E+2012%2E+Automatic+discourse+connective+detection+in+biomedical+text%2E+Journal+of+the+American+Medical+Informatics+Associ)
13. Robert E. Schapire and Yoram Singer. 2000. Boostexter: A boosting-based system for text categorization. Machine Learning, 39(2/3):135–168. [find](https://scholar.google.com/scholar?q=Robert+E%2E+Schapire+and+Yoram+Singer%2E+2000%2E+Boostexter%3A+A+boosting%2Dbased+system+for+text+categorization%2E+Machine+Learning%2C+39%282%2F3%29%3A135%E2%80%93168%2E)
14. Evgeny A. Stepanov and Giuseppe Riccardi. 2013. Comparative evaluation of argument extraction algorithms in discourse relation parsing. In 13th International Conference on Parsing Technologies (IWPT 2013), pages 36–44. [find](https://scholar.google.com/scholar?q=Evgeny+A%2E+Stepanov+and+Giuseppe+Riccardi%2E+2013%2E+Comparative+evaluation+of+argument+extraction+algorithms+in+discourse+relation+parsing%2E+In+13th+International+Conference+on+Parsing+Technologies+%28IWPT+2) [our card](https://md.research.elamur.ai/free/papers/stepanov-2013-comparative-evaluation-argument-extraction-algorithms/)
15. Maite Taboada and William C. Mann. 2006. Applications of rhetorical structure theory. Discourse Studies, (8):567–88. [find](https://scholar.google.com/scholar?q=Maite+Taboada+and+William+C%2E+Mann%2E+2006%2E+Applications+of+rhetorical+structure+theory%2E+Discourse+Studies%2C+%288%29%3A567%E2%80%9388%2E)
16. Bonnie L. Webber, Markus Egg, and Valia Kordoni. 2011. Discourse structure and language technology. Natural Language Engineering, pages 1 – 54. [find](https://scholar.google.com/scholar?q=Bonnie+L%2E+Webber%2C+Markus+Egg%2C+and+Valia+Kordoni%2E+2011%2E+Discourse+structure+and+language+technology%2E+Natural+Language+Engineering%2C+pages+1+%E2%80%93+54%2E)
17. Ben Wellner and James Pustejovsky. 2007. Automatically identifying the arguments of discourse connectives. In Proceedings of the 2007 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning (EMNLP-CoNLL 2007). [find](https://scholar.google.com/scholar?q=Ben+Wellner+and+James+Pustejovsky%2E+2007%2E+Automatically+identifying+the+arguments+of+discourse+connectives%2E+In+Proceedings+of+the+2007+Joint+Conference+on+Empirical+Methods+in+Natural+Language+Processi)
18. Fan Xu, Qiao Ming Zhu, and Guo Dong Zhou. 2012. A unified framework for discourse argument identification via shallow semantic parsing. In Proceedings of 24th International Conference on Computational Linguistics (COLING 2012): Posters. [find](https://scholar.google.com/scholar?q=Fan+Xu%2C+Qiao+Ming+Zhu%2C+and+Guo+Dong+Zhou%2E+2012%2E+A+unified+framework+for+discourse+argument+identification+via+shallow+semantic+parsing%2E+In+Proceedings+of+24th+International+Conference+on+Computational)
19. Hiroyasu Yamada and Yuji Matsumoto. 2003. Statistical dependency analysis with support vector machines. In Proceedings of 8th International Workshop on Parsing Technologies. [find](https://scholar.google.com/scholar?q=Hiroyasu+Yamada+and+Yuji+Matsumoto%2E+2003%2E+Statistical+dependency+analysis+with+support+vector+machines%2E+In+Proceedings+of+8th+International+Workshop+on+Parsing+Technologies%2E) [our card](https://md.research.elamur.ai/free/papers/parsers-2003-yamada-statistical-dependency-analysis-support-vector/)