Made by: Santiago Villarreal Use AI in order to predict the quality of a red wine This project will use the data set of kaggle https://www.kaggle.com/datasets/uciml/red-wine-quality-cortez-et-al-2009
This dataset has 1599 samples, with 12 labels
The project was made using Python and Colab. The code contains multiple tests on the data in order to obtain the best method for analysis. For the develop of the project I follow the next steps
- Import all the libraries needed
- Import the .cvs with all the data and clean the nan values from it
- Made some plots for a visual analisis
- Get the x and y values and impor our clean data
- split the data in 70% 20% and 10% for train validate and test
- standarization
- PCA
10.Train and test KNN and Decision tree with crossvalidation
Results: KNN
Decision tree
CONCLUSION: Our models obtained very low results, since there is a large number of data which, even having the same range, results in a different result, this can be seen in the visual analysis, so the data is found with very little relationship, also as It can be seen in the test and validate results, the model overfits the test data but does not generalize to the rest of the data.
Youtube link:https://youtu.be/uxIVAhF40Fc