For Sentiment Analysis following are languages/ tools are used
- Python
- Spark
- NLTK Natural Language Tool Kit
- Matplotlib for plotting
Dataset is from:
- Yelp Restaurant Reviews
- Curanted Dataset
Naive Bayes Weights to each word based on training dataset
contains(rude) = True negati : positi = 14.3 : 1.0
contains(wonderful) = True positi : negati = 10.9 : 1.0
contains(amazing) = True positi : negati = 7.3 : 1.0
contains(delicious) = True positi : negati = 7.1 : 1.0
contains(loved) = True positi : negati = 6.9 : 1.0
contains(manager) = True negati : positi = 6.4 : 1.0
contains(asked) = True negati : positi = 6.4 : 1.0
contains(manhattan) = True positi : negati = 6.3 : 1.0
contains(authentic) = True positi : negati = 5.6 : 1.0
contains(fantastic) = True positi : negati = 5.3 : 1.0
contains(attentive) = True positi : negati = 5.0 : 1.0
contains(cozy) = True positi : negati = 5.0 : 1.0
contains(meal) = True positi : negati = 4.6 : 1.0