SentiPY is a sentiment ananlysis system based on the SentiDP algorithm and implemented by Python.SentiDP stands for the "Sentiment Drop Point", utilizes the sentiment word strength(like "good" labeled with the value +3) to classify reviews.
what's the Features
1, Sentiment classificaton based on sentiment word strengths;
2, Sentiment rating;
3, Sentiment effectiveness rank;
4, Extracting aspect-opinions;
We need a Stanford NLP service, but its size is more than 300Mb, so we need to download it manually from Dropbox.
1,install corenlp-python(include the Stanford CoreNLP)
wget https://www.dropbox.com/s/21wkispu69cmk7n/corenlp.tar.gz tar xvzf corenlp.tar.gz cd corenlp-python python corenlp/corenlp.py
then the NLP server is running.
install the python package first
1,nlp(to do segment\pos\parse)
from sentipy import nlp server = nlp.StanfordNLP() print nlp.parser1(server,"我在广州")
from sentipy import senti s = senti.senti() s.sentiFLY("新年快乐")
Thanks for the support of Useease.Inc and Stanford NLP GROUP's John.