Adaptive Melodies: A User-Shift Preference Music Recommendation System

Project description

Adaptive Melodies focuses on enhancing music recommendations by addressing the dynamic nature of users' musical preferences. Our goal is to develop a system that not only recognizes but also adapts to shifts in listeners' tastes over time. Utilizing advanced techniques like Time-Decay Collaborative Filtering (TDCF) and Graph Neural Networks (GNNs), we aim to deliver more relevant and adaptive music suggestions.

Table of Contents

• Getting Started
• Installation
• Acknowledgement
• Contact

Getting Started

To run this project locally for development and testing:

1. Clone the repository to your local machine.
2. Navigate to the project directory.

Installation

Follow these steps to set up the project:

1. Install Python and necessary libraries:

pip install -r requirements.txt

2. Run the setup script (not available yet).

3. Start the local server for backend (not available yet).

Acknowledgments

Special thanks to:

• Project Mentor: Tejumade Afonja
• Contributor: Samuel Oyeneye
• Libraries and resources: #nowplaying-RS benchmark data
• Research authors: Asmita Poddar and Eva Zangerle and Yi-Hsuan Yang

Contact

• Project Maintainer: [email protected]

Requirements:

Your project should involve the following components:

• Data Sourcing: Web scraping or any other data sourcing method.
• Data Cleaning and Prep: Data Cleaning, preparation and basic statistics reporting
• Modeling: Base Model, Model Comparison, Hyper-parameter Tuning and monitoring with experiment management
• Model Deployment : Deploy on the web or mobile. You can leverage Google Colab/Streamlit/Huggyface where possible.
• Requirements.txt: A file for all dependecies required

Here is the timeline for your group projects:

• Project Submission Deadline: December 10, 2023
• Presentation Day: December 16, 2023