Using Natural Language Processing to analyze drug reviews perform sentiment analysis on them.

1. Create three directories Dataset, Model_output, word_embeddings
2. Place python file or jupyter notebook in same directory containing above three folders.
3. Inside Dataset folder place dataset files drugsComTrain_raw.tsv and drugsComTest_raw.tsv. (https://archive.ics.uci.edu/ml/datasets/Drug+Review+Dataset+%28Drugs.com%29)
4. Inside Model_output folder create three directories CNN, CNN_GRU, GRU
5. Inside word_embeddings folder place Glove Embeddings. Make sure word embeddings are text (.txt) file or .vec file and not .zip file. (glove.840B.300d - Common Crawl 840 Billion tokens, 2.2Million vocab, cased, 300 dimensional vectors.)
6. Execute the python file or Jupyter notebook in order below: a. Nehal_Ashish_Project.ipynb b. CNN_sentiment.ipynb c. GRU_sentiment.ipynb d. CNN_GRU_sentiment.ipynb