Danceability Histograms
Model Diagram

Inspiration

What makes a popular party song, well, a popular party song? We set out to find correlations across “club” songs, compare them to regular songs, and developed a model to predict whether a song fits the criteria.

How we built it

We started with learning the Spotify API. With it we were able to search for playlists and songs relating to the topic of club hits. We wanted to compare club songs to non-club songs of similar and other genres, so we also created a dataset with curated data from 6 genres, half relevant for clubs and the other half not. With this dataset we were able to make comparisons in tableau and run machine-learning to analyze the components of a club compatible song.

Challenges we ran into

The API data has tags related to different fields so there was a need to carefully plan and go back and forth to retrieve the necessary data.

Accomplishments that we're proud of

Creating our targeted data set and achieving a result with a certain level of accuracy.

What we learned

How to use the API to retrieve relevant data and visualize it in a meaningful way, especially to carefully plan these steps in the context of the project question.