My Capstone Project for the EPFL Applied Data Science: Machine Learning Certificate of Open Studies (COS).
In this project, I use data obtained from Discogs, the largest online second-hand marketplace for physical music, to create a non-individualistic estimation of album market-value for Jazz albums. With this, I aim to identify the divergence between the market-value of albums and the estimation of their value on the basis of their meta-attributes in an attempt to identify to what extent it is effectively possible to predict the true market-value of an album without having listened to the music.
For a more in-depth explanation of the motivation behind this project, please refer to the 00-Introduction.ipynb notebook in this repository.