Modeling Morphology With Linear Discriminative Learning: Considerations and Design Choices

This study addresses a series of methodological questions that arise when modeling inflectional morphology with Linear Discriminative Learning. Taking the semi-productive German noun system as example, we illustrate how decisions made about the representation of form and meaning influence model perf...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Maria Heitmeier, Yu-Ying Chuang, R. Harald Baayen
Formato:	article
Lenguaje:	EN
Publicado:	Frontiers Media S.A. 2021
Materias:	German nouns linear discriminative learning semi-productivity multivariate multiple regression Widrow-Hoff learning frequency of occurrence Psychology BF1-990
Acceso en línea:	https://doaj.org/article/a5c5d9c819bc4eb898546622a6a73999
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:a5c5d9c819bc4eb898546622a6a73999
record_format	dspace
spelling	oai:doaj.org-article:a5c5d9c819bc4eb898546622a6a739992021-11-15T04:44:16ZModeling Morphology With Linear Discriminative Learning: Considerations and Design Choices1664-107810.3389/fpsyg.2021.720713https://doaj.org/article/a5c5d9c819bc4eb898546622a6a739992021-11-01T00:00:00Zhttps://www.frontiersin.org/articles/10.3389/fpsyg.2021.720713/fullhttps://doaj.org/toc/1664-1078This study addresses a series of methodological questions that arise when modeling inflectional morphology with Linear Discriminative Learning. Taking the semi-productive German noun system as example, we illustrate how decisions made about the representation of form and meaning influence model performance. We clarify that for modeling frequency effects in learning, it is essential to make use of incremental learning rather than the end-state of learning. We also discuss how the model can be set up to approximate the learning of inflected words in context. In addition, we illustrate how in this approach the wug task can be modeled. The model provides an excellent memory for known words, but appropriately shows more limited performance for unseen data, in line with the semi-productivity of German noun inflection and generalization performance of native German speakers.Maria HeitmeierYu-Ying ChuangR. Harald BaayenFrontiers Media S.A.articleGerman nounslinear discriminative learningsemi-productivitymultivariate multiple regressionWidrow-Hoff learningfrequency of occurrencePsychologyBF1-990ENFrontiers in Psychology, Vol 12 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	German nouns linear discriminative learning semi-productivity multivariate multiple regression Widrow-Hoff learning frequency of occurrence Psychology BF1-990
spellingShingle	German nouns linear discriminative learning semi-productivity multivariate multiple regression Widrow-Hoff learning frequency of occurrence Psychology BF1-990 Maria Heitmeier Yu-Ying Chuang R. Harald Baayen Modeling Morphology With Linear Discriminative Learning: Considerations and Design Choices
description	This study addresses a series of methodological questions that arise when modeling inflectional morphology with Linear Discriminative Learning. Taking the semi-productive German noun system as example, we illustrate how decisions made about the representation of form and meaning influence model performance. We clarify that for modeling frequency effects in learning, it is essential to make use of incremental learning rather than the end-state of learning. We also discuss how the model can be set up to approximate the learning of inflected words in context. In addition, we illustrate how in this approach the wug task can be modeled. The model provides an excellent memory for known words, but appropriately shows more limited performance for unseen data, in line with the semi-productivity of German noun inflection and generalization performance of native German speakers.
format	article
author	Maria Heitmeier Yu-Ying Chuang R. Harald Baayen
author_facet	Maria Heitmeier Yu-Ying Chuang R. Harald Baayen
author_sort	Maria Heitmeier
title	Modeling Morphology With Linear Discriminative Learning: Considerations and Design Choices
title_short	Modeling Morphology With Linear Discriminative Learning: Considerations and Design Choices
title_full	Modeling Morphology With Linear Discriminative Learning: Considerations and Design Choices
title_fullStr	Modeling Morphology With Linear Discriminative Learning: Considerations and Design Choices
title_full_unstemmed	Modeling Morphology With Linear Discriminative Learning: Considerations and Design Choices
title_sort	modeling morphology with linear discriminative learning: considerations and design choices
publisher	Frontiers Media S.A.
publishDate	2021
url	https://doaj.org/article/a5c5d9c819bc4eb898546622a6a73999
work_keys_str_mv	AT mariaheitmeier modelingmorphologywithlineardiscriminativelearningconsiderationsanddesignchoices AT yuyingchuang modelingmorphologywithlineardiscriminativelearningconsiderationsanddesignchoices AT rharaldbaayen modelingmorphologywithlineardiscriminativelearningconsiderationsanddesignchoices
_version_	1718428823523950592

Modeling Morphology With Linear Discriminative Learning: Considerations and Design Choices

Ejemplares similares