positiveepsilon / GA_Project

Data Science Project
0 stars 0 forks source link

White Wine Quality

Goal: Predict Quality ranking of white wine based on its chemical properties.

"A predictive model developed on this data is expected to provide guidance to vineyards regarding quality and price expected on their produce without heavy reliance on volatility of wine tasters." - Source

Data was obtained from source above. For this projcet we only use the white wine data (Wine_data.csv) provided and not the training and test data which is also available. Instead we create a train_test_split via Scikit-Learn.

Hypothesis: We believe the chemical properties of wine do not contribute to its quality.

Questions:

  1. Which chemical properties play a bigger role in the quality of white wine?

Methods & Models:

  1. Multiple Regression
  2. K-Nearest Neighbors for Quality Classification