• Project Title: Using Machine Learning to Predict Song Popularity

  • BASIS Advisor: Ms. Bhattacharya

  • Internship Location: PG&E

  • Onsite Mentor: Mr. Pranav Jhumkhawala

Music is defined as vocal or instrumental sounds, which contain a set of interrelated dimensions: pitch, duration, dynamics, tempo, timbre, texture and structure. I am using machine-learning algorithms to decipher whether there are trends in music that make certain songs popular over others, and by training my dataset, I will build a song popularity predictor through which future music can be tested in. In my project, working with PG&E (Pacific Gas & Electric),I hope to figure out if hit music (on the Billboard chart) reached its chart position due to an overarching trend in music or if it’s merely luck and the performer’s fame which separate one song from another. This project is significant because if music composers, through machine-learning algorithms, can understand what makes their songs popular, they can utilize the software for future song releases.