A sequence labelling approach for automatic analysis of ello: tagging pronouns, antecedents, and connective phrases

Encapsulators are linguistic units which establish coherent referential connections to the preceding discourse in a text. In this paper, we address challenge of automatically analysing pronominal encapsulator ello Spanish Our method identifies, for each occurrence, antecedent pronoun (including its grammatical type), connective phrase combines with express relation linking text segment following segment, and type semantic expressed by complex marker formed pronoun. We describe our annotation corpus inform development finetune an automatic analyser based on bidirectional encoder representation transformers. On testing method, find that it performs greater accuracy than three baselines (0.76 resolution task), sets promising benchmark occurrences ello, their antecedents, relations between two segments linked combination