Elamur Research⌕
☰
home / papers / 2014 / Representing Caused Motion in Embodied Construction Grammar

Representing Caused Motion in Embodied Construction Grammar

Summary

The paper examines how to formally describe caused-motion events, that is, situations in which an agent causes another object to move. The goal is to enable a system to draw deep semantic inferences from text. The authors combine two resources. The first is FrameNet, in which the frames Motion and Caused_motion are described mostly in prose. The second is Embodied Construction Grammar (ECG), a formalism in which a grammar consists of schemas and constructions, that is, form–meaning pairs. The authors build the CauseMotion schema as a special case of a general cause–effect schema. Its effect is translational motion: MotionPath together with the Source-Path-Goal schema. They also create two argument-structure constructions: one for simple motion and the other for transitive caused motion. The Constructional Analyzer (Bryant 2008) parses a pair of sentences about an envelope sliding into a mailbox. As output, it produces a SemSpec, a network of schemas with bindings between roles and their fillers. The paper contains no quantitative evaluation. The authors show that the analyzer automatically identifies the frame, its roles, and their fillers. Such output can serve as a basis for question answering, information extraction, and event simulation systems.

Key points

  • The paper continues the work of Chang et al. (2002), who formalized the Commerce frame in ECG. The size of FrameNet at that time: almost 1,200 frames, more than 12,740 lexical units, and about 200,000 manually annotated sentences.
  • The CauseMotion schema inherits the cause and affected-entity roles from CauseEffect. The mover role from the MotionPath schema is bound to affected, and the motion itself to effect.
  • The Active_Motion_Path construction consists of a verb and a PP of type Path-PP. Active_Transitive_Caused_Motion additionally has an NP after the verb. In the second construction, the subject is bound to the causer role, so the event is presented from the perspective of the one who causes it.
  • In the SemSpec, the mailbox is understood as a BoundedObject, that is, a container. The starting point of the motion lies outside it, and the endpoint inside.
  • Directions for future work: other motion verbs (roll, slip, fall), interrogative sentences, other frames (perception, communication), and metaphorical motion such as "slide into poverty". Metaphors require constraints on what can fill the roles.

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

Cite

Cite as Ellen K. Dodge and Miriam R. L. Petruck. 2014. Representing Caused Motion in Embodied Construction Grammar. Proceedings of the ACL 2014 Workshop on Semantic Parsing.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.3115/v1/w14-2408

@inproceedings{dodge2014representing,
  title = {{Representing Caused Motion in Embodied Construction Grammar}},
  author = {Ellen K. Dodge and Miriam R. L. Petruck},
  booktitle = {Proceedings of the ACL 2014 Workshop on Semantic Parsing},
  year = {2014},
  doi = {10.3115/v1/w14-2408},
  url = {https://doi.org/10.3115/v1/w14-2408}
}

References

11 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. J. Bryant. 2008. Best-Fit Constructional Analysis. Ph.D. Thesis, University of California, Berkeley. find
  2. N. Chang, S. Narayanan, M. R. L. Petruck. 2002. Putting frames in perspective, Proc. of the 19 th COLING, International Conference on Computational Linguistics. Taipei, Taiwan. find
  3. E. K. Dodge. 2010.Constructional and Conceptual Composition Ph.D. Thesis, University of California, Berkeley. find
  4. J. Feldman, E. Dodge, J. Bryant. 2010. Embodied Construction Grammar. In B. Heine and H. Narrog (eds.), The Oxford Handbook of Linguistic Analysis, Oxford: Oxford University Press, pp. 111-158. find
  5. A. Goldberg. 1995. Constructions: A Construction Grammar Approach to Argument Structure. Chicago: University of Chicago Press. find
  6. N. Friedman, M. Linial, I. Nachman and D. Pe’er. 2000. Using Bayesian networks to analyze expression data. Journal of Computational Biology 7.3-4: 601-620. find
  7. R. Girju. 2003. Automatic detection of causal relations for question answering, Proceedings of the ACL Workshop on Multilingual Summarization and Question Answering: 76-83. find
  8. I. Guyon, A. Elisseeff and C. Aliferis. 2007. Causal feature selection, Computational Methods of Feature Selection Data Mining and Knowledge Discovery Series. Chapman and Hall/CRC, Boca Raton, London, New York, pp. 63-85. find
  9. S. Narayanan. 1997. Knowledge-based Action Representations for Metaphor and Aspect (KARMA). Ph.D. Thesis, University of California, Berkeley. find
  10. Q. Do, Y. Chan, and D. Roth. 2011. Minimally supervised event causality identification, Proceedings of Empirical Methods in Natural Language Processing (EMNLP), Edinburgh, Scotland, UK. find
  11. J. Ruppenhofer, M. Ellsworth, M. R. L. Petruck, C. R. Johnson, and J. Scheffczyk. 2010. FrameNet II: Extended Theory and Practice (Web Publication available via http://framenet.icsi.berkeley.edu). ↗

This page in Markdown → https://md.research.elamur.ai/free/en/papers/dodge-2014-representing-caused-motion-embodied-construction/