Spotify Data Analyst

This program uses Google's Gemini AI and other calculations to get some analytical insight's on a user's Spotify listening data.

Comment

Inspiration

I love seeing the Spotify Wrapped every year, but it makes me want to see more specific data about my listening.

What it does

This project can tell you your top 5 artists, percentage of songs skipped, most common hour of listening, shuffle rate, most skipped artists, most finished artists, and longest listening streak.

How we built it

On Google's Colab (Jupyter Notebook file), I used data I got from Kaggle to analyze with libraries such as pandas and Google Generative AI.

Challenges we ran into

I had a hard time initially figuring out how to connect to the Gemini API as well as how to calculate the most skipped/finished artists.

Accomplishments that we're proud of

I am proud of figuring how to connect to the AI API and getting it to use a prompt I gave it. It was something I've never done before.

What we learned

I learned about using AI in a Python program and how to perform data analysis on a .csv file. I also learned the importance of hiding your API key

What's next for Spotify Data Analyst

I might try to ask Spotify for my own data. Apparently that data comes in a .json formatting so I want to see if I can try to use that and clean my own data to them analyze it. I want to add even more niche analyses.

Built With

Share this project:

Updates