Does the NLTK have any pre-prepared classifiers for Sentiment Analysis?

Question

I compared NLTK and Stanford CoreNLP and found out that the latter had an implementation of RNTN (recursive tensor neural network), designed for mood analysis. Examples available on the Internet show that we don’t need to train him, as he has already been trained using large datasets such as Penn TreeBank

Does NLTK provide a similar feature? The reason I'm asking about this is because any NLTK implementations I found on the Internet include training for a particular class, such as Naive-Bayes or MaxEnt.

PS: Is it just because Python is easy to use, that NLTK is more popular? I'm more comfortable in Java, so I have to choose Stanford CoreNLP or switch to NLTK + python

Answer 1

I found that NLTK already has many pre-processed functions. For example, on the page of this page http://www.nltk.org/api/nltk.sentiment.html you will find a preliminary sentiment analysis configured for social networks (look for "VADER"). If necessary, you can also check this: http://www.nltk.org/nltk_data/ It provides a complete list of all the data sets included in the NLTK, and some through processed data (all kinds of lexicons, such as the famous WORDNET) . I have no experience with Stanford coreNLP, so I can not give any relevant comparison. Anyway, hope this help :)