Nasty.Language.Spanish.CoreferenceConfig (Nasty v0.3.0)

View Source

Configuration for Spanish coreference resolution.

Provides Spanish-specific pronoun lists, gender/number patterns, and resolution rules for the generic coreference resolver.

Spanish Pronoun System

Spanish pronouns have:

  • Gender: masculine/feminine (él/ella)
  • Number: singular/plural (él/ellos)
  • Case: subject/object (él/lo, ella/la)
  • Formality: formal/informal (tú/usted)

Pro-drop

Spanish allows null subjects, so coreference resolution must handle:

  • "Vino ayer" (He/she came yesterday) - no explicit subject
  • Verb conjugation indicates person/number

Example

iex> config = CoreferenceConfig.get()
iex> config.pronouns.subject
["yo", "tú", "él", "ella", "usted", "nosotros", ...]

Summary

Functions

get()

Returns Spanish coreference configuration for use by the generic resolver.

get_gender(map)

Returns the gender of a Spanish token based on morphological features.

get_number(map)

Returns the number of a Spanish token based on morphological features.

pronoun?(token)

Returns true if the given token is a Spanish pronoun.

Functions

get()

@spec get() :: map()

Returns Spanish coreference configuration for use by the generic resolver.

get_gender(map)

@spec get_gender(map()) :: :masculine | :feminine | :unknown

Returns the gender of a Spanish token based on morphological features.

get_number(map)

@spec get_number(map()) :: :singular | :plural | :unknown

Returns the number of a Spanish token based on morphological features.

pronoun?(token)

@spec pronoun?(String.t()) :: boolean()

Returns true if the given token is a Spanish pronoun.