A review of Spanish corpora annotated with negation

Proceedings of the 27th International Conference on Computational Linguistics, pages 915–924 Santa Fe, New Mexico, USA, August 20-26, 2018. 915 A review of Spanish corpora annotated with negation Salud Mar´ıa Jim´enez-Zafra1, Roser Morante2, Mar´ıa Teresa Mart´ın-Valdivia1, L. Alfonso Ure˜na-L´opez1 1 SINAI, Computer Science Department, Advanced Studies Center in ICT (CEATIC) Universidad de Ja´en, Campus Las Lagunillas s/n, E-23071 {sjzafra, maite, laurena}@ujaen.es 2 CLTL Lab, Computational Linguistics VU University Amsterdam, De Boelelaan 1105, 1081 HV r.morantevallejo@vu.nl Abstract The availability of corpora annotated with negation information is essential to develop negation processing systems in any language. However, there is a lack of these corpora even for languages like English, and when there are corpora available they are small and the annotations are not always compatible across corpora. In this paper we review the existing corpora annotated with negation in Spanish with the purpose of first, gathering the information to make it available for other researchers and, second, analyzing how compatible are the corpora and how has the linguistic phenomenon of negation been addressed. Our final aim is to develop a supervised negation processing system for Spanish, for which we need training and test data. Our analysis shows that it will not be possible to merge the small corpora existing for Spanish due to lack of compatibility in the annotations. Title and Abstract in Spanish Revisi´on de los corpus espa˜noles anotados con negaci´on La disponibilidad de corpus anotados con informaci´on sobre la negaci´on es esencial para cualquier idioma, ya que son necesarios para poder desarrollar sistemas capaces de procesar este fen´omeno ling¨u´ıstico. Sin embargo, hay una escasez de corpus anotados con negaci´on, incluso para idiomas como el ingl´es, y cuando hay corpus disponibles, en la mayor´ıa de los casos son peque˜nos y las anotaciones no siempre son compatibles entre ellos. En este trabajo revisamos los corpus anotados con informaci´on sobre la negaci´on en espa˜nol con el prop´osito, en primer lugar, de recopilar la informaci´on para que est´e disponible para otros investigadores y, en segundo lu- gar, de analizar la compatibilidad de los corpus y las estructuras de negaci´on que se han anotado. Nuestro objetivo final es desarrollar un sistema autom´atico para el procesamiento de la negaci´on en espa˜nol. Nuestro an´alisis muestra que no ser´a posible unir los peque˜nos corpus existentes actualmente en espa˜nol debido a la falta de compatibilidad en las anotaciones. 1 Introduction Nowadays, there is a vast amount of information on the Internet. The large number of sources and the high volume of texts make it difficult for users to select information of interest. In order to extract fine- grained information, automatic systems need to be able to process a diversity of linguistic phenomena such as negation, irony or sarcasm that are used to add extra-propositional meaning. In this study, we focus on negation. Negation is a very relevant linguistic phenomenon for most of Natural Language Procesing (NLP) tasks, such as information extraction, question answering and sentiment analysis, since negation cues act as operators that can change the meaning of the …

Content truncated to 500 words