Function stem

pattern/vector/__init__.py:215–243

Returns the base form of the word when counting words in count(). With stemmer=PORTER, the Porter2 stemming algorithm is used. With stemmer=LEMMA, either uses Word.lemma or inflect.singularize(). (with optional parameter language="en", pattern.en.inflect is used).

(word, stemmer=PORTER, **kwargs)

PORTER, LEMMA = "porter", "lemma"
def stem(word, stemmer=PORTER, **kwargs):
    """ Returns the base form of the word when counting words in count().
        With stemmer=PORTER, the Porter2 stemming algorithm is used.
        With stemmer=LEMMA, either uses Word.lemma or inflect.singularize().
        (with optional parameter language="en", pattern.en.inflect is used).
    """
    if isinstance(word, basestring):
        word = decode_utf8(word.lower())
    if stemmer is None:
        return word.lower()
    if stemmer == PORTER:
        return _stemmer.stem(word, **kwargs)
    if stemmer == LEMMA:
        if word.__class__.__name__ == "Word":
            w = word.string.lower()
            if word.lemma is not None:
                return word.lemma
            if word.pos == "NNS":
                return singularize(w)
            if word.pos.startswith(("VB", "MD")):
                return conjugate(w, "infinitive") or w
            if word.pos.startswith(("JJ",)):
                return predicative(w)
            if word.pos.startswith(("DT", "PR", "WP")):
                return singularize(w, pos=pos)
        return singularize(word, pos=kwargs.get("pos", "noun"))
    if hasattr(stemmer, "__call__"):
        return decode_utf8(stemmer(word))
    return word.lower()

def count(words=[], top=None, threshold=0, stemmer=None, exclude=[], stopwords=False, language=None, **kwargs):
    """ Returns a dictionary of (word, count)-items, in lowercase.