Merge pull request #1 from aksh93/master

Sentiment analysis using text-processing.com api

Author: m-arav

Sentiment analysis/sentiment.py

@@ -0,0 +1,25 @@
+#label:	will be either pos if the text is determined to be positive, neg if the text is negative, or neutral if the text is neither pos nor neg.
+#probability: an object that contains the probability for each label. neg and pos will add up to 1, while neutral is standalone. If neutral is greater than 0.5 then the label will be neutral. Otherwise, the label will be pos or neg, whichever has the greater probability.
+
+import json
+import requests
+
+url = 'http://text-processing.com/api/sentiment/'
+
+string = raw_input()
+
+analysis_string = {'text': string}
+
+result = requests.post(url, data=analysis_string)
+
+negative_val = result.json()['probability']['neg']
+print negative_val
+
+positive_val = result.json()['probability']['pos']
+print positive_val
+
+neutral_val = result.json()['probability']['neutral']
+print neutral_val
+
+label = result.json()['label']
+print label