Skip to content

In this project, the idea was to simulate a business problem where a client (winery) had some data about their products and they wanted to see how the variables measured could possibly affect the wine quality, therefore, resulting in higher or lower grades for their wines. The work was divided in 3 main parts: 1. Exploring the data and creating …

Notifications You must be signed in to change notification settings

gurezende/Wine_Quality_Predictions

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Wine_Quality_Predictions

In this project, the idea was to simulate a business problem where a client (winery) had some data about their products and they wanted to see how the variables measured could possibly affect the wine quality, therefore, resulting in higher or lower grades for their wines. Then the client wanted to have their dataset divided in two clusters (white and red wines) and finally see some predictions about the quality for their new wines. The work was divided in 3 main parts:

  1. Exploring the data and creating some data visualizations, understanding the correlations between variables. The result is on the first three tabs of the app: Histograms, Boxplots and Report.
  2. Divide the dataset in two clusters. I have gathered both datasets (red + white) to simulate a request from the client where they needed to know which wine was what kind. The result was 98% accurate and that can be seen on the Clustering tab.
  3. The final part was to create a Prediction Interactive Tool for the White Wines from the dataset that allows the client to input some values and have a prediction whether that wine is going to be qualified as High Quality (grades from 7-10) or Low Quality (grades from 3-6). You can interact with the tool in the tab Predictions. The accuracy is 95%. See more information on this link.

Visit the app here.

About

In this project, the idea was to simulate a business problem where a client (winery) had some data about their products and they wanted to see how the variables measured could possibly affect the wine quality, therefore, resulting in higher or lower grades for their wines. The work was divided in 3 main parts: 1. Exploring the data and creating …

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages