Wine quality, as Maynard Amerine once said, is easier to detect than define. This is due to wine quality being primarily subjective to each individual.
In this project, I use R and apply data analysis techniques to aim to quantify and to relate this very subjective taste of the wine with the chemical properties of the wine, which are more objective.
Exploratory data analysis is an approach to analyzing data sets to summarize their main characteristics, often with visual methods (Wikipedia). The following report details patterns discovery of red wine quality through data exploration and building of the predictive model based on chemical properties.
The complete R markdown can be access here.
The dataset is provided by Udacity and can be downloaded here.
R 3.2.2 or above is required to run the file and the following libraries need to be installed:
- ggplot2
- GGally
- gridExtra
The program is released under a MIT license.