A Statistical Analysis of the famous Wine Reviews Dataset
Specifically, I try to answer if certain wines are better than others? Can the textual reviews be exploited to rate the wines?
Here is a snapshot of EDA:
A wordcloud showing prevalence of words in the textual reviews:
When sentiments are extracted:
Final comparison of the different models:
See the final report for a complete analysis and explanation.