Extraction of Literary Character Information in Portuguese

This chapter describes PALAVRAS-DIP, a sys-tem for the automatic identification of characters and their social profiles in Portuguese and Brazil-ian literature. The system has been designed as an add-on module for a morphosyntactic and semantic parser. We tag human named entities (NE) for pro-fession and social position, and use Constraint Gram -mar (CG relational tags to keep track of co-reference (e.g. pronoun anaphora, zero-subject verbs) and fam-ily reations between the characters. The resulting base annotation allows the extraction of character net-works. The extraction program recognizes and bun-dles character name variants and distinguishes between names with a narrative function and simple cultural references. System development was moti-vated by DIP, a shared-task evaluation on 100 histor-ical novels, where a prototype version achieved rea-sonable F-scores for character identification (63.4%) and alias resolution (68.1%), but underperformed for family relations (15.5%).