---
title: "Combining Manual Rules and Supervised Learning for Hedge"
description: "The paper tackles hedging in biomedical writing: finding the words that weaken a claim (cues) and the… 9 references, cited by 6 papers here."
authors: ["Marek Rei", "Ted Briscoe"]
year: 2010
venue: "Proceedings of the Fourteenth Conference on Computational Natural Language Learning – Shared Task"
source: "https://aclanthology.org/W10-3008/"
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/en/papers/rei-2010-combining-manual-rules-supervised-learning/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/rei-2010-combining-manual-rules-supervised-learning/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2010](https://md.research.elamur.ai/free/en/papers/2010/page/2/) / Combining Manual Rules and Supervised Learning for Hedge Cue and Scope Detection

# Combining Manual Rules and Supervised Learning for Hedge Cue and Scope Detection

**Authors:** [Marek Rei](https://md.research.elamur.ai/free/en/authors/marek-rei/) and [Ted Briscoe](https://md.research.elamur.ai/free/en/authors/ted-briscoe/) · **Year:** 2010 · **Venue:** [Proceedings of the Fourteenth Conference on Computational Natural Language Learning – Shared Task](https://md.research.elamur.ai/free/en/venues/conll/) ([CoNLL 2010](https://md.research.elamur.ai/free/en/venues/conll/2010/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **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/W10-3008/) · [PDF](https://aclanthology.org/W10-3008.pdf) · [OpenAlex](https://openalex.org/W62684482)

**Citations:** 14 citations (OpenAlex, 2026-10-14) · 6 in our library, of the 141,840 papers whose bibliographies we read

## Summary
The paper tackles hedging in biomedical writing: finding the words that weaken a claim (cues) and the stretch of the sentence they weaken (scope). It was built for the CoNLL-2010 shared task, trained on a revised BioScope corpus (9 full papers and 1273 abstracts) and tested on a new set of full papers. Every sentence is parsed with RASP. Cue detection uses a CRF tagger with FILO labels and features drawn from RASP grammatical relations (GRs), lemmas and POS. Its output is then checked against the cues seen in training, and anything unmatched is dropped. For scope, hand-written rules walk the GR graph, and the rule chosen depends on the cue's POS class. A second CRF corrects the rule output, and a few post-processing rules turn the token tags into continuous scopes. With predicted cues, scope F-score climbs from 0.50 (baseline) to 0.53 (rules) to about 0.56 (rules plus CRF). The authors also argue that scopes defined over the surface string are a poor fit for what hedging really affects.

## Key points
- Training data: 2670 sentences from papers and 11,871 from abstracts, with 682 and 2694 cues respectively. Most scopes begin with their cue (72% in papers, about 81% in abstracts).
- Lemma is the most useful feature for cue detection. Unseen cues are handled poorly, so the system keeps only predictions that match a cue from training, using the longest match. This raises precision without hurting recall.
- There are eight scope rules, selected by the cue's generalised POS: modal VM, verb VV, adjective JJ, adverb RR, noun NN, conjunction CC, "whether" CSW, and a default. Graph traversal halts at text-adjunct GRs and clause separators. Passive constructions and raising verbs such as *appear*/*seem* trigger extra rules that pull in the subject.
- Results with predicted cues on 1033 evaluation scopes: baseline F 0.500, manual rules 0.525, rules plus second CRF 0.556 to 0.559. With gold cues: baseline 0.57, rules 0.64.
- Error analysis of the rules: about 35% of wrong scopes come from wrong or fragmented RASP graphs, and about 65% from cases the rules do not cover.

## Interesting ideas
- Hand-written rules make a first guess, and their output is then given to a learned sequence tagger as one more feature, so the learner only has to fix the exceptions instead of learning the whole pattern.
- A learned detector's predictions are filtered against a lexicon built from the training data, dropping anything the model never saw, on the reasoning that the classifier cannot be trusted on unfamiliar items.
- The scope rule is picked by the cue's part of speech, so modals take their verb's subtree, raising verbs and passives add the subject, and conjunctions mark their own subtree. A single small dispatch table captures a lot of syntax.
- Paths through the dependency graph between a word and the cue, plus shorter sub-paths, serve as features, letting a classifier learn graph regularities that the rules express by hand.
- The authors argue that scope belongs over semantic propositions rather than continuous surface spans, and show sentences where keeping the span continuous drags certain claims into a hedged region.

## For Mova

How these ideas could serve Mova, our small open language model — its files, stages and tests:

- **I1** (learn): Self-induced repair rules could feed their verdicts to a trained corrector as features, which keeps every decision traceable to a rule plus a learned adjustment.
- **I2** (frames): Speech-act or modality markers found by a tagger could be checked against a curated list of hedge or attitude words before they change a frame's certainty.
- **I3** (frames): Mova could mark which propositions in the UD tree are hedged by applying rules keyed on the cue's UPOS, then attach an uncertainty flag to the frames inside that subtree.
- **I4** (tree): Bounded dependency-path features could help any token-level decision that hangs on the parse, such as attaching modifiers or negation scope.
- **I5** (world): The world of a text should store certainty per proposition or state, not per text span, so a hedged clause does not weaken a certain one nested inside it.

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

## Institutions

- [University of Cambridge](https://md.research.elamur.ai/free/en/universities/university-of-cambridge/) — University, Cambridge, United Kingdom · 155 works · 115,562 citations

📚 Shelves (4): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 9,085 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 8,888 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 9,296 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 9,206 works

## Cite

Cite as: Marek Rei and Ted Briscoe. 2010. Combining Manual Rules and Supervised Learning for Hedge Cue and Scope Detection. Proceedings of the Fourteenth Conference on Computational Natural Language Learning – Shared Task.

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

```bibtex
@inproceedings{rei2010combining,
  title = {{Combining Manual Rules and Supervised Learning for Hedge Cue and Scope Detection}},
  author = {Marek Rei and Ted Briscoe},
  booktitle = {Proceedings of the Fourteenth Conference on Computational Natural Language Learning – Shared Task},
  year = {2010},
  doi = {10.17863/cam.13157},
  url = {https://doi.org/10.17863/cam.13157}
}
```

- [← previous Presupposition Accommodation as Exception Handling](https://md.research.elamur.ai/free/en/papers/groote-2010-presupposition-accommodation-as-exception-handling/)
- [next → A Global Relaxation Labeling Approach to Coreference Resolution](https://md.research.elamur.ai/free/en/papers/sapena-2010-global-relaxation-labeling-approach-coreference/)
- [↑ back to papers of 2010](https://md.research.elamur.ai/free/en/papers/2010/page/2/)

## References

9 references, 5 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Ted Briscoe, John Carroll, and Rebecca Watson. 2006. [The second release of the RASP system](https://md.research.elamur.ai/free/en/papers/briscoe-2006-second-release-rasp-system/). In Proceedings of the COLING/ACL 2006 on Interactive Presentation Sessions.
2. Richárd Farkas, Veronika Vincze, György Móra, János Csirik, and György Szarvas. 2010. [The CoNLL-2010 Shared Task: Learning to Detect Hedges and their Scope in Natural Language Text](https://md.research.elamur.ai/free/en/papers/farkas-2010-conll-2010-shared-task-learning/). In Proceedings of the Fourteenth Conference on Computational Natural Language Learning (CoNLL-2010): Shared Task, pages 1–12.
3. Halil Kilicoglu and Sabine Bergler. 2008. Recognizing speculative language in biomedical research articles: a linguistically motivated perspective. BMC Bioinformatics, 9 Suppl 11. [find](https://scholar.google.com/scholar?q=Halil+Kilicoglu+and+Sabine+Bergler%2E+2008%2E+Recognizing+speculative+language+in+biomedical+research+articles%3A+a+linguistically+motivated+perspective%2E+BMC+Bioinformatics%2C+9+Suppl+11%2E)
4. John Lafferty, Andrew McCallum, and Fernando Pereira. 2001. Conditional random fields: Probabilistic models for segmenting and labeling sequence data. In Proceedings of ICML 2001. [find](https://scholar.google.com/scholar?q=John+Lafferty%2C+Andrew+McCallum%2C+and+Fernando+Pereira%2E+2001%2E+Conditional+random+fields%3A+Probabilistic+models+for+segmenting+and+labeling+sequence+data%2E+In+Proceedings+of+ICML+2001%2E)
5. Marc Light, Xin Y. Qiu, and Padmini Srinivasan. 2004. The language of bioscience: Facts, speculations, and statements in between. In Proceedings of BioLink 2004. [find](https://scholar.google.com/scholar?q=Marc+Light%2C+Xin+Y%2E+Qiu%2C+and+Padmini+Srinivasan%2E+2004%2E+The+language+of+bioscience%3A+Facts%2C+speculations%2C+and+statements+in+between%2E+In+Proceedings+of+BioLink+2004%2E)
6. Ben Medlock and Ted Briscoe. 2007. [Weakly supervised learning for hedge classification in scientific literature](https://md.research.elamur.ai/free/en/papers/medlock-2007-weakly-supervised-learning-hedge-classification/). In Proceedings of ACL 2007.
7. Roser Morante and Walter Daelemans. 2009. [Learning the scope of hedge cues in biomedical texts](https://md.research.elamur.ai/free/en/papers/morante-2009-learning-scope-hedge-cues-biomedical/). In Proceedings of the Workshop on BioNLP.
8. György Szarvas. 2008. [Hedge classification in biomedical texts with a weakly supervised selection of keywords](https://md.research.elamur.ai/free/en/papers/szarvas-2008-hedge-classification-biomedical-texts-weakly/). In Proceedings of ACL 2008.
9. Veronika Vincze, György Szarvas, Richárd Farkas, György Móra, and János Csirik. 2008. The Bio-Scope corpus: biomedical texts annotated for uncertainty, negation and their scopes. BMC Bioinformatics, 9 Suppl 11. [find](https://scholar.google.com/scholar?q=Veronika+Vincze%2C+Gyo%CC%88rgy+Szarvas%2C+Richa%CC%81rd+Farkas%2C+Gyo%CC%88rgy+Mo%CC%81ra%2C+and+Ja%CC%81nos+Csirik%2E+2008%2E+The+Bio%2DScope+corpus%3A+biomedical+texts+annotated+for+uncertainty%2C+negation+and+their+scopes%2E+BMC+Bioinformati)

*Updated 24 October 2026*