Inspiration
We wanted to learn the fundamentals of NLP and how it is used in the real world.
What it does
The user inputs a song from this dataset. The program reads the description of the given song, ignoring filler words. Then, a linear kernel is built to assess how similar the user's song is compared to every other song in the dataset––it is comparing their song descriptions from a scale 0.0 to 1.0, with a higher score signifying a better match. The program outputs the 10 most similar songs.
How we built it
We coded in Python and used Numpy, Pandas, and Skylearn libraries.
Challenges we ran into
Other datasets were too big to analyze within this hackathon's time frame.
Accomplishments that we're proud of
We learned how and when to implement a linear kernel! This was instrumental to assessing the similarity scores of each song.
What we learned
We learned more about NLP and applied it!
What's next for NLP Song Recommendation
We would like to use a bigger dataset for a wider range of song recommendations.
Log in or sign up for Devpost to join the conversation.