Pregunta

is there a way to show which words are filler words in a given text using NLTK? if not does anyone know where I can get a wordlist with english fill words? thank you

SOLVED: from nltk.corpus import stopwords

¿Fue útil?

Solución

NLTK doesn't provide such a list by itself, though many are available elsewhere on the Web.

There exist quite a number of sources: Web searchs for wordlists with "profanity" "badwords.txt" or blacklists.txt will yield many sources.

In our company's case, we ended up creating our own list and adding to it as needed. Depending on your audience, the list has to be tweaked and adjusted.

Finally, Even though this SO question is closed (and about php) I have found the references and the discussion very useful.

UPDATE: What you want is a list of STOP WORDS.

  1. Try: http://www.ranks.nl/resources/stopwords.html
  2. MIT also maintains a list of stop words.

Hope that helps.

Licenciado bajo: CC-BY-SA con atribución
No afiliado a StackOverflow
scroll top