VOL. 53, BOOK 1, PART A, 2015, pp. 576 – 586 Full text (En)
Author: Svetlozara Leseva, Ivelina Stoyanova, Svetla Koeva
Affiliation: Institute for Bulgarian Language, Bulgarian Academy of Sciences
Abstract
The paper presents a pattern-based method for verb form recognition. We discuss the morphological features of Bulgarian verb forms from the point of view of computational linguistics with a focus on analytical forms, their structure, word order and the possibilities for insertion of external elements between the components of the forms. The patterns defining the verb forms use the lemma, the part of speech and grammatical features. Our future work is focused on improving the method, as well as on integrating the verb form recognition in various NLP applications, such as clause splitting, parallel alignment at various linguistic levels, semantic role labelling, etc.
Key words: pattern-based syntactic analysis, verb form recognition
