pattern.es

The pattern.es module contains a fast part-of-speech tagger for Spanish (identifies nouns, adjectives, verbs, etc. in a sentence) and tools for Spanish verb conjugation and noun singularization & pluralization.

It can be used by itself or with other pattern modules: web | db | en | search | vector | graph.

Documentation

The functions in this module take the same parameters and return the same values as their counterparts in pattern.en. Refer to the documentation there for more details.  

Noun singularization & pluralization

For Spanish nouns there is singularize() and pluralize(). The implementation is slightly less robust than the English version (accuracy 94% for singularization and 78% for pluralization).

>>> from pattern.es import singularize, pluralize
>>>  
>>> print singularize('gatos')
>>> print pluralize('gato')

gato
gatos 

Verb conjugation

For Spanish verbs there is conjugate(), lemma(), lexeme() and tenses(). The lexicon for verb conjugation contains about 600 common Spanish verbs, composed by Fred Jehle. For unknown verbs it will fall back to a rule-based approach with an accuracy of about 84%. 

Spanish verbs have more tenses than English verbs. In particular, the plural differs for each person, and there are additional forms for the FUTURE and CONDITIONAL tense, the IMPERATIVE and SUBJUNCTIVE mood and the PERFECTIVE aspect:

>>> from pattern.es import conjugate
>>> from pattern.es import INFINITIVE, PRESENT, PAST, SG, SUBJUNCTIVE, PERFECTIVE
>>>  
>>> print conjugate('soy', INFINITIVE)
>>> print conjugate('soy', PRESENT, 1, SG, mood=SUBJUNCTIVE)
>>> print conjugate('soy', PAST, 3, SG) 
>>> print conjugate('soy', PAST, 3, SG, aspect=PERFECTIVE) 

ser
sea
era 
fue   

For PAST tense + PERFECTIVE aspect we can also use PRETERITE. For PAST tense + IMPERFECTIVE aspect we can also use IMPERFECT:

>>> from pattern.es import conjugate
>>> from pattern.es import IMPERFECT, PRETERITE
>>>  
>>> print conjugate('soy', IMPERFECT, 3, SG)
>>> print conjugate('soy', PRETERITE, 3, SG)

era
fue   

 The conjugate() function takes the following optional parameters:

Instead of optional parameters, a single short alias, or PARTICIPLE or PAST+PARTICIPLE can also be given. With no parameters, the infinitive form of the verb is returned.

Reference: Jehle, F. (2012). Spanish Verb Forms. Retrieved from: http://users.ipfw.edu/jehle/verblist.htm.

Attributive & predicative adjectives 

Spanish adjectives inflect with an -o, -a , -os, -as, or -es suffix (e.g., curioso → los gatos curiosos) depending on gender. You can get the base form with the predicative() function, or vice versa with attributive(). For predicative, a statistical approach is used with an accuracy of 93%. For attributive, you need to supply gender (MALE, FEMALE, NEUTRAL and/or PLURAL).

>>> from pattern.es import attributive, predicative
>>> from pattern.es import FEMALE, PLURAL 
>>>  
>>> print predicative('curiosos') 
>>> print attributive('curioso', gender=FEMALE)
>>> print attributive('curioso', gender=FEMALE+PLURAL)

curioso
curiosa 
curiosas  

Parser

For parsing there is parse(), parsetree() and split(). The parse() function annotates words in the given string with their part-of-speech tags (e.g., NN for nouns and VB for verbs). The parsetree() function takes a string and returns a tree of nested objects (Text → Sentence → Chunk → Word). The split() function takes the output of parse() and returns a Text. See the pattern.en documentation (here) how to manipulate Text objects. 

>>> from pattern.es import parse, split
>>>  
>>> s = parse('El gato negro se sienta en la estera.')
>>> for sentence in split(s):
>>>     print sentence

Sentence('El/DT/B-NP/O gato/NN/I-NP/O negro/JJ/I-NP/O'
         'se/PRP/B-NP/O sienta/VB/B-VP/O'
         'en/IN/B-PP/B-PNP la/DT/B-NP/I-PNP estera/NN/I-NP/I-PNP ././O/O')

The parser is trained on the Spanish portion of Wikicorpus  using 1.5M words from the tagged sections 10,000–15,000. The accuracy is around 92%. The original Parole tagset is mapped to Penn Treebank tagset. If you need to work with the original tags you can also use parse() with an optional parameter tagset="parole".

Reference: Reese, S., Boleda, G., Cuadros, M., Padró, L., Rigau, G (2010). 
Wikicorpus: A Word-Sense Disambiguated Multilingual Wikipedia Corpus. Proceedings of LREC'10. 

Sentiment analysis

There's no sentiment() function for Spanish yet.