--- title: "Using Hedge Detection to Improve Committed — Ulinski 2018" description: "The paper adds hedge features to a committed-belief tagger. The tagger marks each proposition head as committed belief (CB), non-committed belief (NCB)…" authors: ["Morgan Ulinski", "Seth Benjamin", "Julia Hirschberg"] year: 2018 venue: "Proceedings of the Workshop on Computational Semantics beyond Events and Roles" source: "https://aclanthology.org/W18-1301/" doi: "10.18653/v1/w18-1301" licence: "CC BY 4.0 (ACL Anthology)" licence_url: "https://creativecommons.org/licenses/by/4.0/" summary_licence: "Elamur Cosmic License" canonical: "https://research.elamur.ai/free/en/papers/ulinski-2018-hedge-detection-improve-committed-belief/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/ulinski-2018-hedge-detection-improve-committed-belief/) [home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2018](https://md.research.elamur.ai/free/en/papers/2018/page/4/) / Using Hedge Detection to Improve Committed Belief Tagging # Using Hedge Detection to Improve Committed Belief Tagging **Authors:** [Morgan Ulinski](https://md.research.elamur.ai/free/en/authors/morgan-ulinski/), Seth Benjamin, [Julia Hirschberg](https://md.research.elamur.ai/free/en/authors/julia-hirschberg/) · **Year:** 2018 · **Venue:** Workshop on Computational Semantics beyond Events and Roles (SemBEaR 2018) **Link:** [aclanthology.org/W18-1301](https://aclanthology.org/W18-1301/) · **Licence:** [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/) [**8** citations](https://openalex.org/works?filter=cites:W2807597433) OpenAlex · 2026-10-14 · [**7** citations in our library](https://md.research.elamur.ai/free/en/papers/ulinski-2018-hedge-detection-improve-committed-belief/cited-by/) of the 140673 papers whose bibliographies we read · [Proceedings of the Workshop on Computational Semantics beyond Events and Roles](https://aclanthology.org/volumes/W18-13/) · [licence](https://creativecommons.org/licenses/by/4.0/) - [Publisher's page](https://aclanthology.org/W18-1301/) - [PDF](https://aclanthology.org/W18-1301.pdf) - [DOI 10.18653/v1/w18-1301](https://doi.org/10.18653/v1/w18-1301) - [OpenAlex](https://openalex.org/W2807597433) **Authors (3)** 1. [Morgan Ulinski](https://openalex.org/A5053473331) [Semantic Scholar](https://www.semanticscholar.org/author/3037767) · h 7 · 122 citations · 18 works 2. [Seth Benjamin](https://openalex.org/A5091687445) [Semantic Scholar](https://www.semanticscholar.org/author/2080103439) · h 2 · 826 citations · 3 works 3. [Julia Hirschberg](https://openalex.org/A5045037642) [ORCID](https://orcid.org/0000-0003-0689-7616) · [Semantic Scholar](https://www.semanticscholar.org/author/144049352) · h 67 · 21279 citations · 471 works Citations and author metrics: OpenAlex, 2026-10-14; authors with the same name are not always told apart. ## What it is about The paper adds hedge features to a committed-belief tagger. The tagger marks each proposition head as committed belief (CB), non-committed belief (NCB), reported belief (ROB) or non-belief (NA: desire, intention, question). Hedges are found from a 117-term dictionary in two ways: plain lemma lookup, or hand rules over POS and dependencies that separate hedge from non-hedge uses. For example, *about* as a preposition is not a hedge; *likely* as `amod` of a noun is not; *assume*, *tend* and *appear* are hedges only with a clausal complement; *sure* and *completely* are hedges only under negation; *supposed to* and *should have* are not. Features for the word itself, its parent, children, siblings and ancestors, plus a sentence flag, go into an SVM. On the DEFT forum corpus (5-fold CV), rule-based hedge features raise NCB F1 from 23.7 to 30.4 and ROB from 23.3 to 27.4; overall F1 rises only from 67.5 to 68.1, because those classes are rare. ## Ideas - I1. Give each Mova frame a commitment value next to `neg` and `cond` in `prag::frames` (`prag/src/frames.rs`): committed, non-committed (a hedge or modal governs the predicate), reported (inside the `quote` of a Say frame whose speaker is not the narrator), or not a belief (wish, intention, question). Most of the information is already present: Say frames, `cond`, modals in `feats`. - I2. Let the student weigh by it. In `world/src/knowledge.rs` a remembered frame that is non-committed («maybe the wolf ate it») or reported («the fox said the cheese was poisoned») should support an option less than a committed narrator fact. Measure course test counts and the `world knowledge diagnose` step-5 share (right option outscored). - I3. Hedge rules that look at the parent and siblings, not only the token: a hedge counts for a proposition when it is the head's child (`advmod`) or the head's governing verb («I think [p]»). The same tree position decides which frame inherits non-commitment in I1. Test it on frame bench lines with *think*, *maybe* or *seem*. - I4. Negation flips boosters into hedges («not sure», «isn't completely true»). Reuse `infer::negated` for this one rule in the hedge slot. ## For Mova [languages/en/32-politeness-and-hedging.md](https://mova.elamur.ai/free/en/docs/languages/en/32-politeness-and-hedging/) §8 (epistemic modality and evidentiality), [languages/en/33-implicature-and-presupposition.md](https://mova.elamur.ai/free/en/docs/languages/en/33-implicature-and-presupposition/) (speaker commitment); `prag/src/frames.rs`, `prag/src/feats.rs`, `world/src/knowledge.rs`; measures: frame bench F1 unchanged (no regression), course test counts, the `diagnose` breakdown. ## Cite The paper itself, not this card. BibTeX from the library's record of the work. [DOI 10.18653/v1/w18-1301](https://doi.org/10.18653/v1/w18-1301) ```bibtex @inproceedings{ulinski2018using, title = {{Using Hedge Detection to Improve Committed Belief Tagging}}, author = {Ulinski, Morgan and Benjamin, Seth and Hirschberg, Julia}, booktitle = {Proceedings of the Workshop on Computational Semantics beyond Events and Roles}, year = {2018}, doi = {10.18653/v1/w18-1301}, url = {https://doi.org/10.18653/v1/w18-1301} } ``` - [← previous Using active learning to expand training data for implicit discourse relation recognition](https://md.research.elamur.ai/free/en/papers/xu-2018-active-learning-expand-training-data/) - [next → Using Linguistic Features to Improve the Generalization Capability of Neural Coreference Resolvers](https://md.research.elamur.ai/free/en/papers/moosavi-2018-linguistic-features-improve-generalization-capability/) - [↑ back to papers of 2018](https://md.research.elamur.ai/free/en/papers/2018/page/4/) ## References 4 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet. 1. Veronika Vincze, György Szarvas, Richárd Farkas, György Móra, and János Csirik. 2008. The bioscope corpus: biomedical texts annotated for uncertainty, negation and their scopes. BMC Bioinformatics, 9(11):S9. [↗](https://doi.org/10.1186/1471-2105-9-s11-s9) 2. David Clausen. 2010. Hedgehunter: A system for hedge detection and uncertainty classification. In Proceedings of the Fourteenth Conference on Computational Natural Language Learning, pages 120– 125, Uppsala, Sweden. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=David+Clausen%2E+2010%2E+Hedgehunter%3A+A+system+for+hedge+detection+and+uncertainty+classification%2E+In+Proceedings+of+the+Fourteenth+Conference+on+Computational+Natural+Language+Learning%2C+pages+120%E2%80%93+125%2C+U) 3. Andreas Vlachos and Mark Craven. 2010. Detecting speculative language using syntactic dependencies and logistic regression. In Proceedings of the Fourteenth Conference on Computational Natural Language Learning, pages 18–25, Uppsala, Sweden. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Andreas+Vlachos+and+Mark+Craven%2E+2010%2E+Detecting+speculative+language+using+syntactic+dependencies+and+logistic+regression%2E+In+Proceedings+of+the+Fourteenth+Conference+on+Computational+Natural+Languag) 4. 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. In Proceedings of the Fourteenth Conference on Computational Natural Language Learning, pages 1–12, Uppsala, Sweden. Association for Computational Linguistics. Vinodkumar Prabhakaran, Tomas By, Julia Hirschberg, Owen Rambow, Samira Shaikh, Tomek Strzalkowski, Jennifer Tracey, Michael Arrigo, Ru-payan Basu, Micah Clark, Adam Dalton, Mona [find](https://scholar.google.com/scholar?q=Richa%CC%81rd+Farkas%2C+Veronika+Vincze%2C+Gyo%CC%88rgy+Mo%CC%81ra%2C+Ja%CC%81nos+Csirik%2C+and+Gyo%CC%88rgy+Szarvas%2E+2010%2E+The+conll%2D2010+shared+task%3A+Learning+to+detect+hedges+and+their+scope+in+natural+language+text%2E+In+Proceeding) [our card](https://md.research.elamur.ai/free/en/papers/farkas-2010-conll-2010-shared-task-learning/)