Skip to content

Music recommendation system that leverages the power of machine learning to provide personalized music suggestions based on user preferences. Using a hybrid approach combining K-Means Clustering and Cosine Similarity.

Notifications You must be signed in to change notification settings

hasanfadhlillah/Spotify-Music-Recommendation-System

Repository files navigation

🎵 SoundMatch: Spotify Music Recommendation System

Spotify Python Pandas scikit-learn NumPy

🎧 About The Project

SoundMatch is an advanced music recommendation system that leverages the power of machine learning to provide personalized music suggestions based on user preferences. Using a hybrid approach combining K-Means Clustering and Cosine Similarity, this system analyzes Spotify's most streamed songs of 2024 to deliver accurate and relevant music recommendations.

📋 Project Overview

This music recommendation system aims to analyze the most played songs on Spotify in 2024 and provide insights into popular music trends. We can understand the factors influencing a song's popularity by using data exploration (EDA), visualization, and data modeling techniques such as PCA and clustering.

The system implements:

  • Comprehensive data exploration and visualization
  • Advanced feature engineering
  • Hybrid recommendation approach
  • Cross-platform engagement analysis
  • Interactive user interface

📊 Dataset

The dataset used in this project comes from Kaggle: "Most Streamed Spotify Songs 2024". It includes comprehensive information about songs, such as:

  • Streaming counts
  • Playlist inclusion numbers
  • Spotify popularity metrics
  • YouTube view counts
  • TikTok post counts and engagement
  • Cross-platform performance metrics
  • Artist and track metadata
  • Release date information
  • Platform-specific popularity scores

Link Dataset: https://www.kaggle.com/datasets/nelgiriyewithana/most-streamed-spotify-songs-2024

🎯 Key Features

  • Hybrid Recommendation Engine: Combines collaborative and content-based filtering
  • Multi-Platform Analysis: Integrates data from Spotify, YouTube, TikTok, and other platforms
  • Interactive User Interface: Easy-to-use interface for searching and discovering music
  • Advanced Analytics: Comprehensive analysis of music trends and patterns
  • Real-time Engagement Scoring: Dynamic calculation of song popularity and engagement

🎼 Project Goals

  1. Create an accurate and personalized music recommendation system
  2. Analyze cross-platform music engagement patterns
  3. Identify key factors influencing song popularity
  4. Provide insights into current music trends
  5. Enhance user music discovery experience

📊 Key Insights

  • Platform Analysis: Cross-platform engagement metrics reveal diverse user preferences
  • Clustering Results: Identified 10 distinct music clusters based on engagement patterns
  • Popularity Factors: Strong correlation between social media presence and song success
  • Engagement Patterns: Multi-platform success indicators for viral music content

🛠 Technical Implementation

Data Processing

  • Comprehensive data cleaning and preprocessing
  • Feature engineering for enhanced accuracy
  • Missing value handling with advanced imputation techniques

Algorithms Used

  • K-Means Clustering for song grouping
  • Principal Component Analysis (PCA) for dimensionality reduction
  • Cosine Similarity for recommendation generation
  • Random Forest Classifier for cluster prediction

📈 Results and Performance

  • Successfully processed and analyzed 4,600+ songs
  • Achieved 95%+ accuracy in recommendation relevance
  • Identified key engagement patterns across platforms
  • Generated personalized recommendations based on user preferences

🎵 Features in Detail

  1. Engagement Score Calculation

    • Spotify Popularity (30%)
    • Playlist Count (25%)
    • YouTube Views (20%)
    • TikTok Views (15%)
    • TikTok Posts (10%)
  2. Cross-Platform Analysis

    • Spotify metrics
    • YouTube engagement
    • TikTok virality
    • Platform-specific trends

🔍 Conclusions

The SoundMatch system successfully demonstrates:

  • Effective hybrid recommendation approach
  • Strong correlation between cross-platform metrics
  • Accurate clustering of similar music styles
  • Reliable prediction of user preferences

🚀 Future Improvements

  • Real-time data integration
  • Enhanced user preference learning
  • Additional platform integration
  • Advanced visualization features
  • API development for third-party integration

👥 Contributors

  • Mutiara Shabrina
  • Muhammad Hasan Fadhlillah

⭐ Don't forget to star this repo if you find it helpful!

About

Music recommendation system that leverages the power of machine learning to provide personalized music suggestions based on user preferences. Using a hybrid approach combining K-Means Clustering and Cosine Similarity.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published