This project involves a chemical analysis of wines samples from Italy from three different cultivators, focusing on thirteen different attributes. The goal is to predict the class of wine based on these attributes using K-Nearest Neighbors (K-NN) methods.
We utilize Principal Component Analysis (PCA) to reduce dimensionality, focusing on the first two principal components. Consequently, we apply the K-NN method with varying values of
Data is split into training, validation, and test sets using proportions of 60%, 20%, and 20%, respectively.
UCI ML Wine Recognition Dataset: Wine Dataset
This directory contains some images used in the final report report.pdf