Description This project implements a music recommendation system using K-Means clustering to group similar music preferences. By analyzing features such as danceability, energy, valence, acousticness, loudness, and instrumentalness, the system clusters songs and provides personalized playlist recommendations. Key features include data preprocessing, feature normalization, optimal cluster determination with the elbow method, PCA-based visualization of clusters, animated clustering visualization, and generation of cluster-centric playlists.
Features
- Data Loading and Preprocessing: Handles missing values and retains essential columns.
- Feature Normalization: Standardizes features for consistent clustering.
- K-Means Clustering: Groups songs based on their features.
- Elbow Method Plotting: Determines the optimal number of clusters.
- PCA Visualization: Visualizes clusters in 2D space.
- Animated Clustering Visualization: Shows iterative improvements in clustering.
- Personalized Playlist Recommendation: Recommends songs based on user preferences.
- Cluster-Centric Playlists: Generates playlists for each cluster.