This project delves into Exploratory Data Analysis (EDA) of Spotify data, providing insights into favorite songs, artists, and time spent on the platform. Leveraging these insights, an ongoing binary classification system predicts whether a user will like a new song based on their library.
- Data loading and cleaning of json Spotify data
- Exploration of top artists, songs, genres
- Analysis of listening habits over time
- Binary classification model to predict song likes
- Model uses user's library to determine song suggestions
Building on the EDA, the project extends to a binary classification system for predicting song likability. The user's library becomes the training ground for the classifier, assessing patterns that indicate a user's taste and preferences.
- Review the EDA findings in the respective analysis files.
- Explore the likability binary classifier in progress, residing in a dedicated file.
- Train the classifier on your Spotify library data to predict likability for new songs.
Ensure you have the required libraries installed before running the classifier:
pip install pandas scikit-learnRequest your Spotify data: https://www.spotify.com/us/account/privacy/