We used a dataset from Kaggle that was fetched using Spotify’s API. The dataset consists of 41,106 different songs ranging from 1960-2019 and 19 columns describing each song. Below is a short description of each of the columns as provided by the author of the Kaggle dataset we are using. A more descriptive version of each column can be found at: https://www.kaggle.com/theoverman/the-spotify-hit-predictor-dataset/data.
Published site at: https://aqchan.github.io/Spotify-Hit-Predictor/