added all the files for submission

rohitbhoopalam · Mar 27, 2015 · 2a5fe1d · 2a5fe1d
1 parent 1196e87
commit 2a5fe1d
Show file tree

Hide file tree

Showing 6 changed files with 15,011 additions and 2 deletions.
diff --git a/Help Yelp.pptx b/Help Yelp.pptx
diff --git a/midterm_report.pdf b/midterm_report.pdf
diff --git a/review_15000.json b/review_15000.json
diff --git a/sentimental_analysis.py b/sentimental_analysis.py
@@ -2,12 +2,21 @@
 from nltk.classify import NaiveBayesClassifier
 from nltk.tokenize import word_tokenize 
 from nltk.corpus import stopwords
+from nltk.collocations import BigramCollocationFinder
+from nltk.collocations import BigramAssocMeasures
 import random
+import re
 
 def word_feats(words):
-    return dict([(word, True) for word in words])
+    features = dict([(word, True) for word in words])
+    #finder = BigramCollocationFinder.from_words(words)
+    #finder.apply_word_filter(lambda x: False if re.match('\w', x) else True)
+    #bigrams = finder.nbest(BigramAssocMeasures.chi_sq, 20000)
+    #features.update(dict([(bigram, True) for bigram in bigrams]))
+    return features
 
-f = open('review_small.json', 'r')
+#f = open('review_small.json', 'r')
+f = open('review_15000.json', 'r')
 pos_data = []
 neg_data = []
 

diff --git a/unigram+bigram.png b/unigram+bigram.png
diff --git a/unigram.png b/unigram.png