EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Foundations for MetNet : A Proposed Framework for Modeling Metaphorical Productivity

Ricardo Monteiro, Raquel Amaro

Digital Studies in Language and Literature · 23.11.2026

DOI: 10.1515/dsll-2026-0036 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Digital Studies in Language and Literature, online erschienen am 23.11.2026. Verfasst von Ricardo Monteiro und 1 weiteren Autor:innen. Die Arbeit nennt 75 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

An adequate computational treatment of metaphor requires a representational model that captures its cross-domain, productive nature. Existing approaches and resources fall short: (i) ontological models of general meaning representation, such as WordNet, ConceptNet, BabelNet, and OntoLex, treat concepts atomically, omitting the event structure needed to capture contextual variation; (ii) distributional models (e.g., Word2Vec, BERT) encode metaphorical patterns only implicitly, offering little interpretive insight; (iii) metaphor-specific resources such as MetaNet typically pre-encode mappings rather than deriving them from independent structure, or emphasize ontological (Framester, Amnestic Forgery) or pragmatic (MetCL) representation instead. Although FrameNet – encoding events (frames) and participant roles (frame elements) to support systematic cross-domain alignment – offers a promising foundation, existing approaches built on it leave the underlying linguistic mechanism unaccounted for. In this paper, we review these models and resources to motivate the need for a new theoretical approach to capture this linguistic – and thus systemic – metaphorical mechanism, MetNet , a theoretical framework that refines FrameNet for metaphor modeling through a dedicated reasoning engine, specified through four core tenets: lexical frames, providing the structured, role-based representations over which mappings are established; abstract frames, narrowing the space of plausible candidate pairs; constraints, determining whether a given candidate mapping is admissible; and a mappability scale, grading admitted mappings by the density of their preserved relational structure. Our approach rests on two premises: (i) metaphor should be modeled analogically, as mappings between non-metaphorical frames; and (ii) evaluating such mappings requires a framework capable of encoding graded alignment constraints.

Zitieren

APA

Monteiro, R., & Amaro, R. (2026). Foundations for MetNet : A Proposed Framework for Modeling Metaphorical Productivity. Digital Studies in Language and Literature. https://doi.org/10.1515/dsll-2026-0036

Vancouver

Monteiro R, Amaro R. Foundations for MetNet : A Proposed Framework for Modeling Metaphorical Productivity. Digital Studies in Language and Literature. 2026. doi:10.1515/dsll-2026-0036

BibTeX
@article{Monteiro2026,
  author = {Monteiro, Ricardo and Amaro, Raquel},
  title = {Foundations for MetNet : A Proposed Framework for Modeling Metaphorical Productivity},
  journal = {Digital Studies in Language and Literature},
  year = {2026},
  doi = {10.1515/dsll-2026-0036}
}
Kostenlos starten →