Spotify-Music-Skips-Action-Prediction

Problem Statement:

Predicting whether a user will skip a music during a listening session or not using publicly available data of Spotify.

Web Application (deployment link): https://music-skip-prediction-web-app.herokuapp.com/

ML Algorithm: DecisionTreeClassifier (86% Accurate)

Dataset(s): Publicly available Spotify data. Orginally data is provided in competition hosted on crowdai.org i.e. later moved to crowdai.org. It contain data description file for you to work with problem on your own.

Web application build using: HTML, Flask (Python).

Technologies used: Jupyter Notebook (IDE), Python (Data Scienece/ ML libraries: Pandas, NumPy, Matplotlib, Seaborn, Sklearn, etc), Spyder (IDE)

Methodology:

1. Loading Data
2. Processing Data
3. EDA
4. Feature Engineering
5. ML Model Selection (training & testing)
6. ML Model Deployment (Web App)