Jordi Porta
Departamento de Tecnología y Sistemas, Centro de Estudios de la Real Academia Española, Madrid, Spain
José-Luis Sancho
Departamento de Tecnología y Sistemas, Centro de Estudios de la Real Academia Española, Madrid, Spain
Javier Gómez
Departamento de Tecnología y Sistemas, Centro de Estudios de la Real Academia Española, Madrid, Spain
Download articlePublished in: Proceedings of the workshop on computational historical linguistics at NODALIDA 2013; May 22-24; 2013; Oslo; Norway. NEALT Proceedings Series 18
Linköping Electronic Conference Proceedings 87:6, p. 70-79
NEALT Proceedings Series 18:6, p. 70-79
Published: 2013-05-17
ISBN: 978-91-7519-587-2
ISSN: 1650-3686 (print), 1650-3740 (online)
A system for the analysis of Old Spanish word forms using weighted finite-state transducers is presented. The system uses previously existing resources such as a modern lexicon; a phonological transcriber and a set of rules implementing the evolution of Spanish from the Middle Ages. The results obtained in all datasets show significant improvements; both in accuracy and in the trade-off between precision and recall; with respect to the baseline and the Levenshtein edit distance. A qualitative error analysis suggests several potential ways to improve the performance of the system.
Old Spanish; Finite-State Transducers; Spelling Variation; Historical Linguistics