Inspiration
During the COVID-19 pandemic, connecting with others has been increasingly difficult, with many individuals turning to online resources to communicate. Finding romantic relationships has obviously been no exception. With this in mind, and with the idea that music is the universal language that helps to connect all of us, we decided to create prelude.
What it does
prelude is an online dating app that uses your Spotify data to match you with potential suitors that are also using the app.
How we built it
Spotify rates music tracks based on a handful of factors, including: acousticness, danceability, popularity, tempo, and many more
Use the Spotify API, we average the user's top tracks by relevant metrics, and then match them with other users with similar tastes. For example, Joe’s tracks have an average valence score of 0.3, while Jane’s has an average of 0.4 The difference between their scores is 0.1, so their score would be the inverse, or 10. This score is then summed with the other metrics to create their overall score. Other metrics are also their gender and sexual orientation. On the frontend, we used HTML, CSS, JavaScript, React, React Router. We used Figma to design our components and Firebase to host our application.
Challenges we ran into
We originally decided to use an external library for React components, however we found that the library did not suit our needs a bit too late in the creation process. There were also issues in the workflow where we started designing our components right away, instead of building a wireframe first.
Accomplishments that we're proud of
In terms of frontend, we were able to gain more experience using React, in addition to CSS and React Router. We are also proud of the work we were able to complete in terms of the matching algorithm and the completed pages.
What we learned
We learned that proper planning and prototyping is important, rather than jumping to the creation process immediately.
What's next for prelude
In the future, we would like to add a messaging feature with the users' top matching track playing while they communicate. We also would like to fully utilize our matching algorithm by showing users their matches.
Log in or sign up for Devpost to join the conversation.