Chapter published in:Producing Figurative Expression: Theoretical, experimental and practical perspectives
Edited by John Barnden and Andrew Gargett
[Figurative Thought and Language 10] 2020
► pp. 419–448
Metaphor generation through context sensitive distributional semantics
Stephen McGregor | ction.ai
Matthew Purver | Cognitive Science Research Group, School of Electronic Engineering and Computer Science, Queen Mary University of London, Ljubljana | Department of Knowledge Technologies, InstitutJožef Stefan, Ljubljana
Geraint Wiggins | Cognitive Science Research Group, School of Electronic Engineering and Computer Science, Queen Mary University of London, Ljubljana | Vrije Universiteit Brussel
In this paper, we outline a preliminary methodology for generating metaphor based on contextual projections of representations built up through a statistical analysis of a large-scale linguistic corpus. These projections involve defining subspaces of co-occurrence statistics in which we show that metaphors can be modelled as mappings between congruent regions of semantic representations. We offer this methodology as an empirical implementation pointing towards a resolution of theoretical stances, at times incompatible, construing metaphor as on the one hand an artefact of underlying cognitive processes and on the other hand a product of the environmentally situated generation of ephemeral conceptual schemes.
Keywords: metaphor generation, pragmatics, distributional semantics, natural language processing, context sensitive semantic modelling
Published online: 17 December 2020
Agres, K. R., McGregor, S., Rataj, K., Purver, M., & Wiggins, G. A.
(2016) Modeling metaphor perception with distributional semantics vector space models. In Workshop on Computational Creativity, Concept Invention, and General Intelligence .
Barnden, J. A., & Lee, M. G.
(1999) An implemented context system that combines belief reasoning, metaphor-based reasoning and uncertainty handling. In Modeling and Using Context: Second International and Interdisciplinary Conference (pp. 28– 41).
Baroni, M., & Zamparelli, R.
(2010) Nouns are vectors, adjectives are matrices: Representing adjective-noun constructions in semantic space. In Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing (pp. 1183– 1193).
Barsalou, L. W.
Coecke, B., Sadrzadeh, M., & Clark, S.
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K.
(2019) BERT: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics (pp. 4171–4186).
Ethayarajh, K. (2019) How contextual are contextualized word representations? comparing the geometry of BERT, ELMo, and GPT-2 embeddings. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (pp. 55–65).
Fauconnier, G., & Turner, M. B.
Gargett, A., & J. Barnden (2013) Gen-meta: Generating metaphors using a combination of AI reasoning and corpus-based modeling of formulaic expressions. In Proceedings of TAAI 2013 .
Gibbs Jr., R. W., & Tendahl, M.
Grefenstette, E., Sadrzadeh, M., Clark, S., Coecke, B., & Pulman, S.
Grice, H. P.
Gutiérrez, E. D., Shutova, E., Marghetis, T., & Bergen, B. K.
Kartsaklis, D., & Sadrzadeh, M.
(2013) Prior disambiguation of word tensors for constructing sentence vectors. In Proceedings of the 2013 Conference on Empirical Methods in Natural Language Processing (pp. 1590–1601).
Levinson, S. C.
Mason, Z. J.
McGregor, S., Agres, K., Purver, M., & Wiggins, G.
McGregor, S., Jezek, E., Purver, M., & Wiggins, G.
(2017) A geometric method for detecting semantic coercion. In Proceedings of 12th International Workshop on Computational Semantics .
Mikolov, T., Chen, K., Corrado, G., & Dean, J. (2013) Efficient estimation of word representations in vector space. In Proceedings of ICLR Workshop .
Miyazawa, A., & Miyao, Y.
(2017) Evaluation metrics for automatically generated metaphorical expressions. In 12th International Workshop on Computational Semantics .
Peters, M., Neumann, M., Iyyer, M., Gardner, M., Clark, C., Lee, K., & Zettlemoyer, L.
(2018) Deep contextualized word representations. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics (pp. 2227–2237).
Salton, G., Wong, A., & Yang, C. S.
(1975) A vector space model for automatic indexing. In Proceedings of the 12th ACM SIGIR Conference (pp. 137–150).
(1992) Dimensions of meaning. In Proceedings of the 1992 ACM/IEEE conference on Supercomputing (pp. 787–796).
Searle, J. R.
Shutova, E., Teufel, S., & Korhonen, A.
Steen, G. J., Dorst, A. G., Herrmann, J. B., Kaal, A., Krennmayr, T., & Pasma, T.
Tsvetkov, Y., Boytsov, L., Gershman, A., Nyberg, E., & Dyer, C.
van Genabith, J.
(1999) Metaphors and type theory. In Proceedings of the AISB’99 Symposium on Metaphor, Artificial Intelligence, and Cognition (pp. 108–112).
Veale, T., & Hao, Y.
(2007) Comprehending and generating apt metaphors: A web-driven, case-based approach to figurative language. AAAI , 1471–1476.