EDA on IMDB Dataset from Hugging Face

1. Installed datasets using the below command in python from Jupyter Notebook

pip install datasets

2. Inspected the dataset using dataset_builder

3. Loaded IMDB Dataset for Analysis

4. Investigated Data distribution across Split

5. Calculated counts of positive & negative reviews across splits

6. Compared Length of reviews across splits & review type

7. Tokenized Text for Sentiment Analysis

8. Removed stopwords (from nltk + sklearn + punctuations) from Tokenized word lists

9. Plotted most common words from both positive & negative reviews