Skip to content

A recommendation system that fetches the most similar track to the user's chosen track written in Python

Notifications You must be signed in to change notification settings

anu-bazar/Spotify-Recommendation-ML

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

17 Commits
 
 
 
 

Repository files navigation

Multi-faceted Analysis and Classification of Spotify Tracks

In this project, I utilized the "Spotify Tracks Dataset" available on public domain at Spotify Tracks Dataset. 😊✨

I used standard machine learning techniques such as data exploration and visualization, followed by statistical methods such as doing a PCA and a correlation heatmap analysis. Afterwards, I created a recommendation system that fetches the most similar track to the user's chosen track. Additionally, I compared classification methods based on audio features provided by the dataset such as Random Forest, Gradient boosting, FNN and MLP. I concluded the analysis by analysing one property - "popularity", in order to find what makes a track popular and what influences this. Lastly, I did a genre-based analysis by finding the most energetic, most danceable genres and genres with the highest valence. 🎛️🎶

For this project, I've used the following libraries: Pandas, Sklearn, Numpy, Tensorflow, Keras, Plotly, Seaborn, Matplotlib.

I also created a report on the findings of an in depth analysis to accompany the code. The code is available on a Google Colab notebook and a Python script.

Important

To view the full project, I'd personally suggest you to please view the Jupyter notebook titled spotify.ipynb file for the code with commentary and visualizations for the best experience.

About

A recommendation system that fetches the most similar track to the user's chosen track written in Python

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published