python - NLTK NaiveBayesClassifier training -


For example, I'm following online, the only way to train a given classifier is to do it well Bad tweets on the list of good lists

I think training on only negative and positive words will provide a lot of data and therefore more accurate results are negative than many examples of positive and negative tweets And a list of positive words are very easy to meet.

In that case, you can use single words instead of sentences.

In this way, it will fit according to the example you mentioned. Neg_tweets = [('Hate', 'Positive') ('Bad') ('Angry', 'Positive'), ('Good', 'Positive'), 'Negative']] Pre>


Comments

Popular posts from this blog

import - Python ImportError: No module named wmi -

Editing Python Class in Shell and SQLAlchemy -

c# - MySQL Parameterized Select Query joining tables issue -