Cover not available
Article published In:
Explorations of morphological structure in distributional space
Edited by Melanie J. Bell, Juhani Järvikivi and Vito Pirrelli
[The Mental Lexicon 17:3] 2022
► pp. 368–394

A generating model for Finnish nominal inflection using distributional semantics

ORCID logo | University of Eastern Finland
| University of Tübingen
| University of Tübingen
Finnish nouns are characterized by rich inflectional variation, with obligatory marking of case and number, with optional possessive suffixes and with the possibility of further cliticization. We present a model for the conceptualization of Finnish inflected nouns, using pre-compiled fasttext embeddings (300-dimensional semantic vectors that approximate words’ meanings). Instead of deriving the semantic vector of an inflected word from another word in its paradigm, we propose that an inflected word is conceptualized by means of summation of latent vectors representing the meanings of its lexeme and its inflectional features. We tested this model on the 2,000 most frequent Finnish nouns and their inflected word forms from a corpus of Finnish (84 million tokens). Visualization of the semantic space of Finnish using t-SNE clarified that a ‘main effects’ additive model does not do justice to the semantics of inflection. In Finnish, how number is realized turns out to vary substantially with case. Further interactions emerged with the possessive suffixes and the clitics. By taking these interactions into account, the accuracy of our model, evaluated with the fasttext embeddings as gold standard, improved from 76% to 89%. Analyses of the errors made by the model clarified that 7.5% of errors are due to overabundance (and hence not true errors), and that 16.5% of the errors involved exchanges of semantically highly similar stems (lexemes). Our results indicate, first, that the semantics of Finnish noun inflection are more intricate than assumed thus far, and second, that these intricacies can be captured with surprisingly high accuracy by a simple generating model based on imputed semantic vectors for lexemes, inflectional features, and interactions of inflectional features.
Keywords: word embeddings, inflectional morphology, fasttext, word2vec, tSNE, imputed semantic vectors, Finnish language
Article outline
  • 1.Introduction
  • 2.Finnish noun inflection
  • 3.Fasttext-based models of Finnish noun semantics
  • 4.A generating model for nominal conceptualization
  • 5.Error analysis
  • 6.Models based on word2vec instead of on fasttext
  • 7.Discussion
  • Notes
  • References
Available under the Creative Commons Attribution (CC BY) 4.0 license.

For any use beyond this license, please contact the publisher at [email protected].

Published online: 17 March 2023
DOI logo
https://doi.org/10.1075/ml.22008.nik
References (34)
References
Baayen, R. H., Chuang, Y.-Y., Shafaei-Bajestan, E., and Blevins, J. (2019). The discriminative lexicon: A unified computational model for the lexicon and lexical processing in comprehension and production grounded not in (de)composition but in linear discriminative learning. Complexity. DOI logoGoogle Scholar
Blevins, J. P. (2016). Word and paradigm morphology. Oxford University Press. DOI logoGoogle Scholar
Boleda, G. (2020). Distributional Semantics and Linguistic Theory. Annual Review of Linguistics, 61:213–234. DOI logoGoogle Scholar
Booij, G. E. (1996). Inherent versus contextual inflection and the split morphology hypothesis. In Booij, G. E. and Marle, J. V., editors, Yearbook of Morphology 1995, pages 1–16. Kluwer Academic Publishers, Dordrecht. DOI logoGoogle Scholar
Brunila, M. and LaViolette, J. (2022). What company do words keep? revisiting the distributional semantics of jr firth & zellig harris. arXiv preprint arXiv:2205.07750.Google Scholar
Bybee, J. L. (1985). Morphology: A study of the relation between meaning and form. Benjamins, Amsterdam. DOI logoGoogle Scholar
Chen, J. and Chen, Z. (2008). Extended bayesian information criteria for model selection with large model spaces. Biometrika, 95(3):759–771. DOI logoGoogle Scholar
Chuang, Y. Y., Brown, D., Evans, R. and Baayen, R. H. (2022). Paradigm gaps are associated with weird “distributional semantics”. Russian defective nouns and their case and number paradigms.Google Scholar
Epskamp, S., Borsboom, D., and Fried, E. I. (2018). Estimating psychological networks and their accuracy: A tutorial paper. Behavior research methods, 50(1):195–212. DOI logoGoogle Scholar
Epskamp, S., Cramer, A. O., Waldorp, L. J., Schmittmann, V. D., and Borsboom, D. (2012). qgraph: Network visualizations of relationships in psychometric data. Journal of statistical software, 481:1–18. DOI logoGoogle Scholar
Firth, J. R. (1968). Selected papers of J R Firth, 1952–59. Indiana University Press.Google Scholar
Grave, E., Bojanowski, P., Gupta, P., Joulin, A., and Mikolov, T. (2018). Learning word vectors for 157 languages. arXiv preprint arXiv:1802.06893.Google Scholar
Günther, F., Rinaldi, L., and Marelli, M. (2019). Vector-Space Models of Semantic Representation From a Cognitive Perspective: A Discussion of Common Misconceptions. Perspectives on Psychological Science, 14(6):1006–1033. DOI logoGoogle Scholar
Harris, Z. S. (1954). Distributional Structure. WORD, 10(2–3). DOI logoGoogle Scholar
Karlsson, F. (1983). Suomen kielen äänne-ja muotorakenne [the phonological and morphological structure of finnish]. Werner Söderström, Juva.Google Scholar
(1985). Paradigms and word forms. Studia gramatyczne, 71:135–154.Google Scholar
(1986). Frequency considerations in morphology. STUF-Language Typology and Universals, 39(1–4):19–28. DOI logoGoogle Scholar
(2017). Finnish: A comprehensive grammar. Routledge. DOI logoGoogle Scholar
Karlsson, F. and Koskenniemi, K. (1985). A process model of morphology and lexicon. Folia Linguistica, 291:207–231. DOI logoGoogle Scholar
Krijthe, J. H. (2015). Rtsne: T-Distributed Stochastic Neighbor Embedding using Barnes-Hut Implementation. R package version 0.16.Google Scholar
Laine, M., Kujala, P., Niemi, J., and Uusipaikka, E. (1992). On the nature of naming difficulties in aphasia. Cortex, 28(4):537–554. DOI logoGoogle Scholar
Landauer, T. and Dumais, S. (1997). A solution to Plato’s problem: The latent semantic analysis theory of acquisition, induction and representation of knowledge. Psychological Review, 104(2):211–240. DOI logoGoogle Scholar
Marelli, M. and Baroni, M. (2015). Affixation in semantic space: Modeling morpheme meanings with compositional distributional semantics. Psychological Review, 122(3):485–515. DOI logoGoogle Scholar
Mikolov, T., Chen, K., Corrado, G., and Dean, J. (2013). Efficient estimation of word representations in vector space. 1st International Conference on Learning Representations, ICLR 2013 – Workshop Track Proceedings, pages 1–12.Google Scholar
Nikolaev, A., Ashaie, S., Hallikainen, M., Hänninen, T., Higby, E., Hyun, J., Lehtonen, M., and Soininen, H. (2019). Effects of morphological family on word recognition in normal aging, mild cognitive impairment, and alzheimer’s disease. Cortex, 1161:91–103. DOI logoGoogle Scholar
Schreuder, R. and Baayen, R. H. (1997). How complex simplex words can be. Journal of Memory and Language, 371:118–139. DOI logoGoogle Scholar
Shafaei-Bajestan, E., Moradipour-Tari, M., Uhrig, P., and Baayen, R. H. (2022). Semantic properties of english nominal pluralization: Insights from word embeddings. arXiv.Google Scholar
Shafaei-Bajestan, Elnaz, Uhrig, Peter and Baayen, R. H. (2023). Making sense of spoken plurals. DOI logoGoogle Scholar
Shahmohammadi, H., Lensch, H., and Baayen, R. H. (2021). Learning zero-shot multifaceted visually grounded word embeddings via multi-task training. CoNLL 2021. arXiv preprint arXiv:2104.07500. DOI logoGoogle Scholar
Sinclair, J. (1991). Corpus, concordance, collocation. Describing English language. Oxford University Press, Oxford.Google Scholar
Tibshirani, R. (1996). Regression shrinkage and selection via the lasso. Journal of the Royal Statistical Society: Series B (Methodological), 58(1):267–288.Google Scholar
van der Maaten, L. (2014). Accelerating t-sne using tree-based algorithms. Journal of Machine Learning Research, 151:3221–3245.Google Scholar
van der Maaten, L. and Hinton, G. (2008). Visualizing high-dimensional data using t-sne. Journal of Machine Learning Research, 91:2579–2605.Google Scholar
Wang, B., Wang, A., Chen, F., Wang, Y., and Kuo, C. C. (2019). Evaluating word embedding models: Methods and experimental results. APSIPA Transactions on Signal and Information Processing, 81(May):e19. DOI logoGoogle Scholar
Cited by (6)

Cited by six other publications

Heikkilä, Timo T., Nea Soralinna & Jukka Hyönä
2024. Relating foveal and parafoveal processing efficiency with word-level parameters in text reading. Journal of Memory and Language 137 ► pp. 104516 ff. DOI logo
Herce, Borja & Marc Allassonnière-Tang
2024. The meaning of morphomes: distributional semantics of Spanish stem alternations. Linguistics Vanguard 0:0 DOI logo
Mujezinović, Erdin, Vsevolod Kapatsinski & Ruben van de Vijver
2024. One Cue's Loss Is Another Cue's Gain—Learning Morphophonology Through Unlearning. Cognitive Science 48:5 DOI logo
van de Vijver, Ruben, Emmanuel Uwambayinema & Yu-Ying Chuang
2024. Comprehension and production of Kinyarwanda verbs in the Discriminative Lexicon. Linguistics 62:1 ► pp. 79 ff. DOI logo
Heitmeier, Maria, Yu-Ying Chuang & R. Harald Baayen
2023. How trial-to-trial learning shapes mappings in the mental lexicon: Modelling lexical decision with linear discriminative learning. Cognitive Psychology 146 ► pp. 101598 ff. DOI logo
Kivisaari, Sasa L., Annika Hultén, Marijn van Vliet, Tiina Lindh-Knuutila & Riitta Salmelin
2023. Semantic feature norms: a cross-method and cross-language comparison. Behavior Research Methods DOI logo

This list is based on CrossRef data as of 5 july 2024. Please note that it may not be complete. Sources presented here have been supplied by the respective publishers. Any errors therein should be reported to them.