noqqe » blog | sammelsurium | photos | projects | about

Sentiment Analysis - Ist der Text Positiv oder negativ?

Words per Sentence

len(nltk.corpus.brown.words()) / len(nltk.corpus.brown.sents())

Anzahl der Wörter finden

len(nltk.tokenize.sent_tokenize(foo))

Anzahl der Sätze finden

len(nltk.tokenize.sent_tokenize(text1))

Search text for every occurence of word:

text1.concordance('foo')

Find synonyms:

text1.similar('foo')

Generate random text:

text1.generate()

Count Vocabulary:

len(text1)

Count occurance of word:

text1.count('foo')

See all words from text:

set(text1)

Hapaxes (least common words):

FreqDist(text1).hepaxes()

Find Long Words:

sorted([w for w in set(text1) in len(w) > 15])

Train Classifier