Preference semantics, ill-formedness, and metaphor
Computational Linguistics - Special issue on ill-formed input
Computational mechanisms for metaphor in languages: a survey
Journal of Computer Science and Technology
A structured vector space model for word meaning in context
EMNLP '08 Proceedings of the Conference on Empirical Methods in Natural Language Processing
Language models based on semantic composition
EMNLP '09 Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing: Volume 1 - Volume 1
From frequency to meaning: vector space models of semantics
Journal of Artificial Intelligence Research
A regression model of adjective-noun compositionality in distributional semantics
GEMS '10 Proceedings of the 2010 Workshop on GEometrical Models of Natural Language Semantics
EMNLP '10 Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing
Distributional memory: A general framework for corpus-based semantics
Computational Linguistics
| Hi-index | 0.00 |
In this paper, we present a first attempt to characterize the semantic deviance of composite expressions in distributional semantics. Specifically, we look for properties of adjective-noun combinations within a vector-based semantic space that might cue their lack of meaning. We evaluate four different compositionality models shown to have various levels of success in representing the meaning of AN pairs: the simple additive and multiplicative models of Mitchell and Lapata (2008), and the linear-map-based models of Guevara (2010) and Baroni and Zamparelli (2010). For each model, we generate composite vectors for a set of AN combinations unattested in the source corpus and which have been deemed either acceptable or semantically deviant. We then compute measures that might cue semantic anomaly, and compare each model's results for the two classes of ANs. Our study shows that simple, unsupervised cues can indeed significantly tell unattested but acceptable ANs apart from impossible, or deviant, ANs, and that the simple additive and multiplicative models are the most effective in this task.