Lexical Function Identification Using Word Embeddings and Deep Learning
In this work, we report the results of our experiments on the task of distinguishing the semantics of verb-noun collocations in a Spanish corpus. This semantics was represented by four lexical functions of the Meaning-Text Theory. Each lexical function specifies a certain universal semantic concept found in any natural language. Knowledge of collocation and its semantic content is important for natural language processing, as collocation comprises the restrictions on how words can be used together. We experimented with a combination of GloVe word embeddings as a recent and extended algorithm for vector representation of words and a deep neural architecture, in order to recover most of the context of verb-noun collocations in a meaningful way which could discriminate among lexical functions. Our corpus was a collection of 1,131 Excelsior newspaper issues. As our results showed, the proposed deep neural architecture outperformed state-of-the-art supervised learning methods.
KeywordsWord embeddings Deep learning Lexical function Meaning-Text Theory
The research was done under partial support of Mexican Government: SNI, BEIFI-IPN, and SIP-IPN grants 20196021, 20196437.
- 1.Enikeeva, E., Popov, A.: Developing a Russian database of regular semantic relations based on word embeddings. In: Krek, S., Cibej, J., Gorjanc, V., Kosem, I. (eds.) Proceedings of the XVIII EURALEX International Congress: Lexicography in Global Contexts, Ljubljana, Slovenia, pp. 799–809 (2018)Google Scholar
- 2.Fonseca, A., Sadat, F., Lareau, F.: Retrieving information from the French lexical network in RDF/OWL format. In: Calzolari, N., et al. (eds.) Proceedings of the 11th International Conference on Language Resources and Evaluation, Miyazaki, Japan (2018)Google Scholar
- 3.Gelbukh, A., Kolesnikova, O.: Supervised learning for semantic classification of Spanish collocations. In: Martínez-Trinidad, J.F., Carrasco-Ochoa, J.A., Kittler, J. (eds.) MCPR 2010. LNCS, vol. 6256, pp. 362–371. Springer, Heidelberg (2010). https://doi.org/10.1007/978-3-642-15992-3_38. The complete list of 737 Spanish verb-noun collocations annotated with 36 lexical functions can be accessed at http://22.214.171.124/okolesnikova/index.php?id=lex/ or http://www.gelbukh.com/lexical-functionsCrossRefGoogle Scholar
- 5.Lux-Pogodalla, V., Polguère, A.: Construction of a French lexical network: methodological issues. In: First International Workshop on Lexical Resources, Ljubljana, Slovenia, pp. 54–61 (2011)Google Scholar
- 6.Mel’čuk, I.A.: Lexical functions: A tool for the description of lexical relations in a lexicon. In: Wanner, L. (ed.) Lexical Functions in Lexicography and Natural Language Processing, pp. 37–102. Benjamins Academic Publishers, Amsterdam (1996)Google Scholar
- 7.Miller, G.A., Leacock, C., Tengi, R., Bunker, R.T.: A semantic concordance. In: Proceedings of the Workshop on Human Language Technology Association for Computational Linguistics, Stroudsburg, PA, pp. 303–308 (1993)Google Scholar
- 8.Mikolov, T., Sutskever, I., Chen, K., Corrado, G.S., Dean, J.: Distributed representations of words and phrases and their compositionality. In: Advances in Neural Information Processing Systems, pp. 3111–3119. Neural Information Processing Systems Foundation, Inc. (2013)Google Scholar
- 9.Polguère A.: Towards a theoretically-motivated general public dictionary of semantic derivations and collocations for French. In: Proceedings of EURALEX 2000, Stuttgart, Germany, pp. 517–527 (2000)Google Scholar
- 10.Tutin, A.: Annotating lexical functions in corpora: showing collocations in context. In: Apresjan, Y., Iomdin, L. (eds.) Proceedings of the Second International Conference on the Meaning-Text Model, pp. 498–510. Slavic Culture Languages Publishing House, Moscow (2007)Google Scholar
- 13.Wanner, L., Bohnet, B., Giereth, M.: What is beyond collocations? Insights from machine learning experiments. In: Proceedings of the 12th EURALEX International Congress, Turin, Italy, pp. 1071–1084 (2006)Google Scholar