BaSCo: An Annotated Basque-Spanish Code-Switching Corpus for Natural Language Understanding

Proceedings of the 13th Conference on Language Resources and Evaluation (LREC 2022), pages 3158–3163 Marseille, 20-25 June 2022 © European Language Resources Association (ELRA), licensed under CC-BY-NC-4.0 3158 BaSCo: An Annotated Basque-Spanish Code-Switching Corpus for Natural Language Understanding Maia Aguirre, Laura Garc´ıa-Sardi˜na, Manex Serras, Ariane M´endez, Jacobo L´opez Vicomtech Foundation, Basque Research and Technology Alliance (BRTA) Parque Cient´ıfico y Tecnol´ogico de Gipuzkoa, Paseo Mikeletegi 57, Donostia / San Sebasti´an (Spain) {magirre, lgarcias, mserras, amendez, jlopez}@vicomtech.org Abstract The main objective of this work is the elaboration and public release of BaSCo, the first corpus with annotated linguistic resources encompassing Basque-Spanish code-switching. The mixture of Basque and Spanish languages within the same utterance is popularly referred to as Euska˜nol, a widespread phenomenon among bilingual speakers in the Basque Country. Thus, this corpus has been created to meet the demand of annotated linguistic resources in Euska˜nol in research areas such as multilingual dialogue systems. The presented resource is the result of translating to Euska˜nol a compilation of texts in Basque and Spanish that were used for training the Natural Language Understanding (NLU) models of several task-oriented bilingual chatbots. Those chatbots were meant to answer specific questions associated with the administration, fiscal, and transport domains. In addition, they had the transverse potential to answer to greetings, requests for help, and chit-chat questions asked to chatbots. BaSCo is a compendium of 1377 tagged utterances with every sample annotated at three levels: (i) NLU semantic labels, considering intents and entities, (ii) code-switching proportion, and (iii) domain of origin. Keywords: Code-Switching, Language-Mixing, Corpus Compilation, Semantic Annotation, Language Proportion, Natural Lan- guage Understanding, Dialogue Systems 1. Introduction The phenomenon of code-switching –i.e., alternating words belonging to two or more different languages or “codes” at utterance boundaries (intersentential code-switch) or within the same utterance (intrasentential code-switch) (Zirker, 2007)– is very common in both oral and written interactions in diglossic areas such as the Basque Country (Barredo, 1997). Basque is an isolated European language that orig- inates from and is spoken mainly in the Basque Country, where currently 32% of its population is bilingual1. A common feature of natural interactions among such bilin- gual speakers is the spontaneous continuous switching be- tween the Basque and Spanish languages. This commu- nicative strategy goes by the popular name of Euska˜nol, which is a blending of the words “Euskara” (Basque) and “Espa˜nol” (Spanish). Despite the widespread universality of code-switching, an- notated linguistic resources encompassing code alterna- tions are very scarce (Hamed et al., 2017). For the case of Euska˜nol, since it is a quite narrowly localised phe- nomenon, the situation is even harsher. To the extent of our knowledge, only a few research papers have been pub- lished on this matter: Lantto (2014) discusses the relation- ship between swearing and slang to Basque-Spanish code- switching. Ezeizabarrena and Aeby (2010) study the syn- tax of Euska˜nol based on observations of interaction sit- uations between children and adults. However, both arti- cles approach code-switching from the standpoint of lin- guistic analysis and do not explore …

Content truncated to 500 words