It's recommended to read the notebooks in the following order:
- data_exploration.ipynb - Here we explore the data and prepare it for further analyses. The data used in this project is taken from Kaggle and contains around 130k wine reviews. Note that this repo doesn't include the data itself, so access it on Kaggle if you need it.
- wine_value_analysis.ipynb - Here we find value wines and their main characteristics. In particular, we explore which countries, regions and grape varieties we should pay attention to in our search for value wines.
- blind_tasting_red_wine.ipynb - Here we build a model that predicts a red wine's grape variety based on the tasting description - this is what sommeliers call "blind tasting". Besides making the prediction, we display the main descriptors, or key words, characteristic of the variety.