Natural language Processing - Programming Assignment

Dataset:

1. Choose one category from http://jmcauley.ucsd.edu/data/amazon/ - amazon product review data.
2. Choose at least 25,000 (reviews). [if no. of reviews > 25k)
3. Review rule, for dataset:
   1. [overall > 3.0] - positive
   2. [overall <= 3.0] - negative

Module - 2 (Sentiment Analysis using statistical NLP):

Tasks:-

1. Use the following vector space models
   1. CountVectorizer.
   2. TF-IDF.
   3. Any external vectorizer (cite the original paper).
2. Do sentiment analysis using all (a,b,c) using classical ML techniques
   1. Naive Bayes Model.
   2. Decision Tree.
   3. Logistic Regression.
3. Report metrics [accuracy, f1 score, confusion matrix] for all the combinations in (1 and 2)
4. Analyse the results. [Report clearly which vector space model is giving better results on each model used]