Analysis of Chat Text

chat_sentiment_analysis.py

# preliminaries
from nltk.sentiment.vader import SentimentIntensityAnalyzer
from nltk.stem.porter import PorterStemmer
from nltk import tokenize, pos_tag
import csv
import re

# read in data
comments = csv.reader(open("/Users/scott/Downloads/issue_comments.csv", "rb"))

# fire up porter stemmer and sentimment analyzer
st = PorterStemmer()
sid = SentimentIntensityAnalyzer()

def get_mentions(line):
	return re.findall('@{1}\\S+', line)

def tokenize_lines(lines):
	return tokenize.sent_tokenize(lines)

def tokenize_sentence(line):
	return [tokenize.word_tokenize(sent) for sent in sentences]

def stem_text(text):
	return ' '.join([st.stem(stem) for stem in text.split(' ')])

def get_polarity(line):
	return sid.polarity_scores(line)


' '.join([st.stem(stem) for stem in foo[5].split(' ')])
st.stem(foo[5])
get_mentions(foo[5])