Unsupervised Machine Learning Project

This is the group project I did for my unsupervised machine learning course. We were the group “Alcoholics” and, as named, we did our project on wine quality. We used the famous wine quality dataset and wanted to see if there could be a good predictor of wine quality based on the content of the wine. We implemented kmeans and spectral clustering with and without PCA for dimensionality reduction and compared it with classification results from supervised machine learning methods. We found that the unsupervised ML methods we used was not a good predictor and that the supervised ML methods were better, but not really good. We talk about our results and why we believe so in this paper, if you are interested: https://docs.google.com/document/d/1lfhR6DyR1tyuVdl2RNEWx_FCF-6LvJyeb2cND3empUU/edit?usp=sharing.

We also have a video version of our report here: https://www.youtube.com/watch?v=7JkbbrmWkEE