---
title: "A Method for Automatically Constructing Case Frames for English"
description: "The paper builds a large lexicon of semantics-oriented case frames for English verbs. 88.4% accuracy. 23 references, cited by 2 papers here."
authors: ["Daisuke Kawahara", "Kiyotaka Uchimoto"]
year: 2008
venue: "Proceedings of the Sixth International Conference on Language Resources and Evaluation (LREC'08)"
source: "https://aclanthology.org/L08-1098/"
licence: "CC BY-NC 4.0 (ELRA, LREC proceedings)"
licence_url: "https://creativecommons.org/licenses/by-nc/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/kawahara-2008-method-automatically-constructing-case-frames/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/kawahara-2008-method-automatically-constructing-case-frames/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2008](https://md.research.elamur.ai/free/en/papers/2008/page/2/) / A Method for Automatically Constructing Case Frames for English

# A Method for Automatically Constructing Case Frames for English

**Authors:** [Daisuke Kawahara](https://md.research.elamur.ai/free/en/authors/daisuke-kawahara/) and [Kiyotaka Uchimoto](https://md.research.elamur.ai/free/en/authors/kiyotaka-uchimoto/) · **Year:** 2008 · **Venue:** [Proceedings of the Sixth International Conference on Language Resources and Evaluation (LREC'08)](https://md.research.elamur.ai/free/en/venues/lrec/) ([LREC 2008](https://md.research.elamur.ai/free/en/venues/lrec/2008/)) · **Publisher:** [European Language Resources Association (ELRA)](https://md.research.elamur.ai/free/en/publishers/european-language-resources-association-elra/) · **Licence:** [CC BY-NC 4.0 (ELRA, LREC proceedings)](https://creativecommons.org/licenses/by-nc/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/L08-1098/) · [PDF](https://aclanthology.org/L08-1098.pdf) · [OpenAlex](https://openalex.org/W205832485)

**Citations:** 3 citations (OpenAlex, 2026-10-14) · 2 in our library, of the 149,439 papers whose bibliographies we read

## Summary
The paper builds a large lexicon of semantics-oriented case frames for English verbs. Each frame records which argument slots a verb takes and which nouns typically fill them, with frequencies. Subcategorization lexicons capture only syntactic patterns, so they lump together different senses of a verb. The authors instead derive frames from automatic parses of a raw corpus.

From the English Gigaword, they keep only short sentences, tag them, and parse them with MSTParser. They then take predicate-argument structures only from sentences that have a single active main verb. These structures are grouped by a dominant argument (object first, then subject, then a prepositional phrase). The resulting groups are merged by clustering that weighs both slot-pattern overlap and WordNet-based word similarity.

The output covers about 9,300 verbs, with 5.5 frames per verb on average. A manual check of 20 random verbs found 88.4% of frames correct.

The resource matters because it gives selectional preferences at the level of observed words, which parsers, translation and paraphrasing systems can use directly.

## Key points
- Source data: English Gigaword Second Edition (LDC2005T12), about 100M newswire sentences. Restricting to sentences of 20 words or fewer left roughly 47M.
- Pipeline:
  - Tsuruoka's tagger for tagging.
  - MSTParser trained on WSJ sections 2–21 of the Penn Treebank.
  - On section 23, the parser reaches 89.9% labeled dependency accuracy and 36.3% complete sentences.
  - On short test sentences, these rise to 91.5% and 56.4%.
- Slots are surface cases (sbj, obj, obj2, pp:in, pp:at, …). Fillers are lemmatized, lowercased headwords with frequencies, not semantic classes.
- Clustering uses the similarity measure of Kawahara and Kurohashi (2006), with word similarity computed by WordNet::Similarity.
- Result: about 9,300 verb entries at 5.5 frames per verb, with 88.4% accuracy on 20 hand-checked verbs.
  - The main errors came from choosing the wrong dominant argument.
  - Some errors came from clustering, which did not treat WordNet synsets.

## Interesting ideas
- Trading recall for precision by harvesting only from inputs where the parser is reliable (short sentences, one active main verb) means corpus size compensates for discarded data. This filtering approach suits any knowledge mined from noisy automatic analyses.
- Seeding sense distinctions by a fixed priority of dominant slot (object > subject > PP) is a cheap, explainable first cut at verb sense splitting: "borrow money" and "borrow an idea" land in separate frames before any clustering happens.
- Using raw filler words with counts rather than a fixed semantic class inventory sidesteps the problem of picking a generalization level. Generalization is deferred to similarity at use time.
- Merging clusters by combining slot-pattern similarity with lexical similarity keeps frames consistent in both syntax and meaning, rather than in one dimension alone.
- The evaluation applies explicit per-frame criteria: no mixed senses, obligatory slots present, and stray fillers tolerated only in non-dominant slots. This makes manual judgments of an induced lexicon repeatable.

## For Mova

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

- **I1** (learn): Silver data agreed by witness parsers could be further restricted to short single-clause sentences before verb knowledge is extracted, keeping the induced lexicon clean.
- **I2** (frames): Frame selection in prag could key candidate senses on the head of the object slot, giving a traceable reason why one reading was chosen over another.
- **I3** (world): Compiled verb knowledge could store frequency-weighted observed fillers per slot, with WordNet similarity applied on demand to judge whether a new entity plausibly fills a role.
- **I4** (frames): When aligning parsed clauses with VerbNet frames, scoring both argument-pattern match and filler similarity could reduce metaphor-driven misassignments.
- **I5** (learn): Similar checklists could serve as a rubric for LLM teacher tiers grading self-induced frames, treating a contaminated dominant slot as a hard failure.

*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

- [National Institute of Information and Communications Technology](https://md.research.elamur.ai/free/en/labs/national-institute-of-information-and-communications-technology/) — Research institute, Tokyo, Japan · 23 works · 2,090 citations

📚 Shelves (4): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works

🧭 Topics (1): [English](https://md.research.elamur.ai/free/en/topics/english/) · 2,263 works

## Cite

Cite as: Daisuke Kawahara and Kiyotaka Uchimoto. 2008. A Method for Automatically Constructing Case Frames for English. Proceedings of the Sixth International Conference on Language Resources and Evaluation (LREC'08).

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

```bibtex
@inproceedings{kawahara2008method,
  title = {{A Method for Automatically Constructing Case Frames for English}},
  author = {Daisuke Kawahara and Kiyotaka Uchimoto},
  booktitle = {Proceedings of the Sixth International Conference on Language Resources and Evaluation (LREC'08)},
  year = {2008},
  doi = {10.63317/4q5b9fuga22p},
  url = {https://doi.org/10.63317/4q5b9fuga22p}
}
```

- [← previous Textual Entailment as an Evaluation Framework for Metaphor Resolution: A Proposal](https://md.research.elamur.ai/free/en/papers/agerri-2008-textual-entailment-as-evaluation-framework/)
- [next → Old English Ge- and the Descriptive Power of Nerthus](https://md.research.elamur.ai/free/en/papers/arista-2008-old-english-ge-descriptive-power/)
- [↑ back to papers of 2008](https://md.research.elamur.ai/free/en/papers/2008/page/2/)

## References

23 references, 8 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Collin Baker, Charles J. Fillmore, and John Lowe. 1998. [The Berkeley FrameNet Project](https://md.research.elamur.ai/free/en/papers/baker-1998-berkeley-framenet/). In Proceedings of COLING-ACL1998, pages 86–90.
2. Bran Boguraev, Ted Briscoe, John Carroll, David Carter, and Claire Grover. 1987. [The derivation of a grammatically-indexed lexicon from the Longman Dictionary of Contemporary English](https://md.research.elamur.ai/free/en/papers/boguraev-1987-derivation-grammatically-indexed-lexicon-longman/). In Proceedings of the 25th Annual Meeting of the Association for Computational Linguistics, pages 193–200.
3. Michael Brent. 1993. From grammar to lexicon: Unsupervised learning of lexical syntax. Computational Linguistics, 19(2):243–262. [find](https://scholar.google.com/scholar?q=Michael+Brent%2E+1993%2E+From+grammar+to+lexicon%3A+Unsupervised+learning+of+lexical+syntax%2E+Computational+Linguistics%2C+19%282%29%3A243%E2%80%93262%2E)
4. Ted Briscoe and John Carroll. 1997. [Automatic extraction of subcategorization from corpora](https://md.research.elamur.ai/free/en/papers/briscoe-1997-automatic-extraction-subcategorization-corpora/). In Proceedings of the 5th Conference on Applied Natural Language Processing, pages 356–363. [↗](https://doi.org/10.48550/arxiv.cmp-lg/9702002)
5. Glenn Carroll and Mats Rooth. 1998. Valence induction with a head-lexicalized PCFG. In Proceedings of the 3rd Conference on Empirical Methods in Natural Language Processing, pages 36–45. [↗](https://doi.org/10.48550/arxiv.cmp-lg/9805001)
6. Michael Ellsworth and Adam Janin. 2007. [Mutaphrase: Paraphrasing with framenet](https://md.research.elamur.ai/free/en/papers/ellsworth-2007-mutaphrase-paraphrasing-framenet/). In Proceedings of the ACL-PASCAL Workshop on Textual Entailment and Paraphrasing, pages 143–150.
7. Murat Ersan and Eugene Charniak, 1996. A Statistical Syntactic Disambiguation Program and What It Learns, pages 146–157. Springer. [find](https://scholar.google.com/scholar?q=Murat+Ersan+and+Eugene+Charniak%2C+1996%2E+A+Statistical+Syntactic+Disambiguation+Program+and+What+It+Learns%2C+pages+146%E2%80%93157%2E+Springer%2E)
8. Susanne Gahl. 1998. Automatic extraction of subcorpora based on subcategorization frames from a part-of-speech tagged corpus. In Proceedings of the 36th Annual Meeting of the Association for Computational Linguistics and 17th International Conference on Computational Linguistics, pages 428–432. [find](https://scholar.google.com/scholar?q=Susanne+Gahl%2E+1998%2E+Automatic+extraction+of+subcorpora+based+on+subcategorization+frames+from+a+part%2Dof%2Dspeech+tagged+corpus%2E+In+Proceedings+of+the+36th+Annual+Meeting+of+the+Association+for+Computati)
9. Ralph Grishman, Catherine Macleod, and Adam Meyers. 1994. [COMLEX Syntax: Building a computational lexicon](https://md.research.elamur.ai/free/en/papers/grishman-1994-comlex-syntax-building-computational-lexicon/). In Proceedings of COLING1994, pages 268–272. [↗](https://doi.org/10.48550/arxiv.cmp-lg/9411017)
10. Masahiko Haruno. 1995. A case frame learning method for Japanese polysemous verbs. In Proceedings of the AAAI Spring Symposium: Representation and Acquisition of Lexical Knowledge: Polysemy, Ambiguity, and Generativity, pages 45–50. [find](https://scholar.google.com/scholar?q=Masahiko+Haruno%2E+1995%2E+A+case+frame+learning+method+for+Japanese+polysemous+verbs%2E+In+Proceedings+of+the+AAAI+Spring+Symposium%3A+Representation+and+Acquisition+of+Lexical+Knowledge%3A+Polysemy%2C+Ambiguity)
11. Daisuke Kawahara and Sadao Kurohashi. 2006. Case frame compilation from the web using high-performance computing. In Proceedings of LREC2006. [find](https://scholar.google.com/scholar?q=Daisuke+Kawahara+and+Sadao+Kurohashi%2E+2006%2E+Case+frame+compilation+from+the+web+using+high%2Dperformance+computing%2E+In+Proceedings+of+LREC2006%2E)
12. Anna Korhonen and Judita Preiss. 2003. Improving subcategorization acquisition using word sense disambiguation. In Proceedings of ACL2003, pages 48–55. [find](https://scholar.google.com/scholar?q=Anna+Korhonen+and+Judita+Preiss%2E+2003%2E+Improving+subcategorization+acquisition+using+word+sense+disambiguation%2E+In+Proceedings+of+ACL2003%2C+pages+48%E2%80%9355%2E)
13. Anna Korhonen, Yuval Krymolowski, and Ted Briscoe. 2006. A large subcategorization lexicon for natural language processing applications. In Proceedings of LREC2006. [find](https://scholar.google.com/scholar?q=Anna+Korhonen%2C+Yuval+Krymolowski%2C+and+Ted+Briscoe%2E+2006%2E+A+large+subcategorization+lexicon+for+natural+language+processing+applications%2E+In+Proceedings+of+LREC2006%2E)
14. Anna Korhonen. 2002. Semantically motivated subcategorization acquisition. In Proceedings of the ACL-02 Workshop on Unsupervised Lexical Acquisition, pages 51–58. [find](https://scholar.google.com/scholar?q=Anna+Korhonen%2E+2002%2E+Semantically+motivated+subcategorization+acquisition%2E+In+Proceedings+of+the+ACL%2D02+Workshop+on+Unsupervised+Lexical+Acquisition%2C+pages+51%E2%80%9358%2E)
15. Maria Lapata. 1999. Acquiring lexical generalizations from corpora: A case study for diathesis alternations. In Proceedings of ACL1999, pages 397–404. [find](https://scholar.google.com/scholar?q=Maria+Lapata%2E+1999%2E+Acquiring+lexical+generalizations+from+corpora%3A+A+case+study+for+diathesis+alternations%2E+In+Proceedings+of+ACL1999%2C+pages+397%E2%80%93404%2E)
16. Christopher Manning and Hinrich Schütze. 1999. Foundations of Statistical Natural Language Processing. The MIT Press, Cambridge, Massachusetts. [find](https://scholar.google.com/scholar?q=Christopher+Manning+and+Hinrich+Schu%CC%88tze%2E+1999%2E+Foundations+of+Statistical+Natural+Language+Processing%2E+The+MIT+Press%2C+Cambridge%2C+Massachusetts%2E)
17. Christopher Manning. 1993. Automatic acquisition of a large subcategorization dictionary from corpora. In Proceedings of the 31st Annual Meeting of the Association for Computational Linguistics, pages 235–242. [find](https://scholar.google.com/scholar?q=Christopher+Manning%2E+1993%2E+Automatic+acquisition+of+a+large+subcategorization+dictionary+from+corpora%2E+In+Proceedings+of+the+31st+Annual+Meeting+of+the+Association+for+Computational+Linguistics%2C+pages)
18. Ryan McDonald, Kevin Lerman, and Fernando Pereira. 2006. [Multilingual dependency analysis with a two-stage discriminative parser](https://md.research.elamur.ai/free/en/papers/mcdonald-2006-multilingual-dependency-analysis-two-stage/). In Proceedings of CoNLL-X, pages 216–220.
19. The XTAG Research Group, editor. 1998. [A Lexicalized Tree Adjoining Grammar for English](https://md.research.elamur.ai/free/en/papers/group-1998-lexicalized-tree-adjoining-grammar-english/).
20. Yoshimasa Tsuruoka and Jun’ichi Tsujii. 2005. [Bidirectional inference with the easiest-first strategy for tagging sequence data](https://md.research.elamur.ai/free/en/papers/tsuruoka-2005-bidirectional-inference-easiest-first-strategy/). In Proceedings of HLT-EMNLP2005, pages 467–474.
21. Akira Ushioda, David Evans, Ted Gibson, and Alex Waibel. 1993. The automatic acquisition of frequencies of verb subcategorization frames from tagged corpora. In Proceedings of the Workshop on Acquisition of Lexical Knowledge from Text, pages 95–106. [↗](https://doi.org/10.5445/ir/1000166599)
22. Takehito Utsuro, Takashi Miyata, and Yuji Matsumoto. 1997. Maximum entropy model learning of subcategorization preference. In Proceedings of the 5th Workshop on Very Large Corpora, pages 246–260. [find](https://scholar.google.com/scholar?q=Takehito+Utsuro%2C+Takashi+Miyata%2C+and+Yuji+Matsumoto%2E+1997%2E+Maximum+entropy+model+learning+of+subcategorization+preference%2E+In+Proceedings+of+the+5th+Workshop+on+Very+Large+Corpora%2C+pages+246%E2%80%93260%2E)
23. Daniel Zeman. 2002. Can subcategorisation help a statistical dependency parser? In Proceedings of COL-ING2002, pages 1156–1162. [find](https://scholar.google.com/scholar?q=Daniel+Zeman%2E+2002%2E+Can+subcategorisation+help+a+statistical+dependency+parser%3F+In+Proceedings+of+COL%2DING2002%2C+pages+1156%E2%80%931162%2E)

*Updated 1 November 2026*