Tags

Type your tag names separated by a space and hit enter

Spicy Adjectives and Nominal Donkeys: Capturing Semantic Deviance Using Compositionality in Distributional Spaces.
Cogn Sci 2017; 41(1):102-136CS

Abstract

Sophisticated senator and legislative onion. Whether or not you have ever heard of these things, we all have some intuition that one of them makes much less sense than the other. In this paper, we introduce a large dataset of human judgments about novel adjective-noun phrases. We use these data to test an approach to semantic deviance based on phrase representations derived with compositional distributional semantic methods, that is, methods that derive word meanings from contextual information, and approximate phrase meanings by combining word meanings. We present several simple measures extracted from distributional representations of words and phrases, and we show that they have a significant impact on predicting the acceptability of novel adjective-noun phrases even when a number of alternative measures classically employed in studies of compound processing and bigram plausibility are taken into account. Our results show that the extent to which an attributive adjective alters the distributional representation of the noun is the most significant factor in modeling the distinction between acceptable and deviant phrases. Our study extends current applications of compositional distributional semantic methods to linguistically and cognitively interesting problems, and it offers a new, quantitatively precise approach to the challenge of predicting when humans will find novel linguistic expressions acceptable and when they will not.

Authors+Show Affiliations

Computer Laboratory, University of Cambridge. Center for Mind/Brain Sciences, University of Trento.Center for Mind/Brain Sciences, University of Trento.Department of Psychology and Cognitive Science, University of Trento.Department of Information Engineering and Computer Science, University of Trento.

Pub Type(s)

Journal Article
Research Support, Non-U.S. Gov't

Language

eng

PubMed ID

26991668

Citation

Vecchi, Eva M., et al. "Spicy Adjectives and Nominal Donkeys: Capturing Semantic Deviance Using Compositionality in Distributional Spaces." Cognitive Science, vol. 41, no. 1, 2017, pp. 102-136.
Vecchi EM, Marelli M, Zamparelli R, et al. Spicy Adjectives and Nominal Donkeys: Capturing Semantic Deviance Using Compositionality in Distributional Spaces. Cogn Sci. 2017;41(1):102-136.
Vecchi, E. M., Marelli, M., Zamparelli, R., & Baroni, M. (2017). Spicy Adjectives and Nominal Donkeys: Capturing Semantic Deviance Using Compositionality in Distributional Spaces. Cognitive Science, 41(1), pp. 102-136. doi:10.1111/cogs.12330.
Vecchi EM, et al. Spicy Adjectives and Nominal Donkeys: Capturing Semantic Deviance Using Compositionality in Distributional Spaces. Cogn Sci. 2017;41(1):102-136. PubMed PMID: 26991668.
* Article titles in AMA citation format should be in sentence-case
TY - JOUR T1 - Spicy Adjectives and Nominal Donkeys: Capturing Semantic Deviance Using Compositionality in Distributional Spaces. AU - Vecchi,Eva M, AU - Marelli,Marco, AU - Zamparelli,Roberto, AU - Baroni,Marco, Y1 - 2016/03/16/ PY - 2013/04/16/received PY - 2015/09/30/revised PY - 2015/10/01/accepted PY - 2016/3/19/pubmed PY - 2017/12/22/medline PY - 2016/3/19/entrez KW - Compositionality KW - Distributional models KW - Meaning representation KW - Semantic deviance KW - Semantic spaces SP - 102 EP - 136 JF - Cognitive science JO - Cogn Sci VL - 41 IS - 1 N2 - Sophisticated senator and legislative onion. Whether or not you have ever heard of these things, we all have some intuition that one of them makes much less sense than the other. In this paper, we introduce a large dataset of human judgments about novel adjective-noun phrases. We use these data to test an approach to semantic deviance based on phrase representations derived with compositional distributional semantic methods, that is, methods that derive word meanings from contextual information, and approximate phrase meanings by combining word meanings. We present several simple measures extracted from distributional representations of words and phrases, and we show that they have a significant impact on predicting the acceptability of novel adjective-noun phrases even when a number of alternative measures classically employed in studies of compound processing and bigram plausibility are taken into account. Our results show that the extent to which an attributive adjective alters the distributional representation of the noun is the most significant factor in modeling the distinction between acceptable and deviant phrases. Our study extends current applications of compositional distributional semantic methods to linguistically and cognitively interesting problems, and it offers a new, quantitatively precise approach to the challenge of predicting when humans will find novel linguistic expressions acceptable and when they will not. SN - 1551-6709 UR - https://www.unboundmedicine.com/medline/citation/26991668/Spicy_Adjectives_and_Nominal_Donkeys:_Capturing_Semantic_Deviance_Using_Compositionality_in_Distributional_Spaces_ L2 - https://doi.org/10.1111/cogs.12330 DB - PRIME DP - Unbound Medicine ER -