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

View Source

Recognizes named entities in Spanish text.

Delegates to generic rule-based entity recognition with Spanish-specific configuration.

Uses rule-based patterns to identify:

  • PERSON: names (Juan García, María López)
  • LOCATION: cities, countries (Madrid, España, Barcelona, Cataluña)
  • ORGANIZATION: companies, institutions (Banco de España, Real Madrid)
  • DATE: temporal expressions (lunes, 15 de enero, 2024)
  • MONEY: currency amounts (100 euros, $50, 25€)
  • PERCENT: percentages (25%, 3.5 por ciento)

Spanish-Specific Features

  • Spanish name lexicons (common Spanish names, surnames)
  • Spanish place lexicons (Spanish cities, regions, Latin American countries)
  • Spanish titles (Sr., Sra., Dr., Dra., Don, Doña)
  • Spanish date formats (15 de enero de 2024)
  • Euro currency symbols (€)
  • Spanish organizational patterns (S.A., S.L., Ltda.)

Example

iex> {:ok, entities} = EntityRecognizer.recognize(spanish_tokens)
{:ok, [%Entity{type: :PERSON, text: "Juan García"}, ...]}

Summary

Functions

recognize(tokens, opts \\ [])

Recognizes named entities in Spanish tokens.

Functions

recognize(tokens, opts \\ [])

@spec recognize(
  [Nasty.AST.Token.t()],
  keyword()
) :: {:ok, [Nasty.AST.Entity.t()]} | {:error, term()}

Recognizes named entities in Spanish tokens.

Delegates to the Spanish adapter which uses generic rule-based entity recognition with Spanish-specific configuration (lexicons, patterns, heuristics).

Options

  • :types - List of entity types to extract (default: all)
  • :min_confidence - Minimum confidence threshold (default: 0.5)
  • :use_context - Use context for disambiguation (default: true)

Examples

iex> {:ok, entities} = EntityRecognizer.recognize(tokens)
{:ok, [%Entity{type: :PERSON, text: "María García"}, ...]}

iex> {:ok, entities} = EntityRecognizer.recognize(tokens, types: [:PERSON, :ORG])
{:ok, [%Entity{...}]}