This repository contains a notebook for a sentiment analysis using the NLTK VADER sentiment analyzer to classify a subset of Amazon e-commerce reviews. Additionally, the relationship between sentiment scoring and the Amazon review ratings system was explored.
Data source from Kaggle: Amazon Reviews Dataset.
The following plot showcases that the VADER scoring system accurately reflects Positive, Neutral, and Negative sentiments for reviews where a 5-star review is considered to be very positive and a 1-star review is considered to be very negative.