How kai-zen23/elective-1 works
This repository is a course-lab analysis of Spotify user behavior, not a streaming-platform implementation. The principal workflow explores a supplied dataset, prepares behavioral features, evaluates and fits K-Means, then visualizes and profiles the resulting user clusters. The notebook also proposes recommendation and UX adaptations based on those profiles; these are described as an application, not implemented integrations. The root README is a team portfolio. The analysis source is sampled only in part, so the graph omits uncertain internal wiring.
Main entry points: FirstLab_CS412.py Fit K-Means · spotify_user_behavior_realistic_50000_rows.xlsx Spotify behavior data
Components 9
Data Analysis
- Spotify behavior data
FirstLab_CS412/spotify_user_behavior_realistic_50000_rows.xlsx - Explore data
FirstLab_CS412/FirstLab_CS412.py - Prepare features
FirstLab_CS412/FirstLab_CS412.py - Choose cluster count
FirstLab_CS412/FirstLab_CS412.py
Clustering Results
- Fit K-Means
FirstLab_CS412/FirstLab_CS412.py - Visualize clusters
FirstLab_CS412/FirstLab_CS412.py - Summarize user profiles
FirstLab_CS412/FirstLab_CS412.py
Proposed Application
- Recommendation and UX proposals
FirstLab_CS412/FirstLab_CS412.py
Other components
- Lab analyst
Connections 9
- Lab analyst → Explore data: runs analysis
- Spotify behavior data → Explore data: provides data
- Explore data → Prepare features: informs selection
- Prepare features → Choose cluster count: evaluates K
- Choose cluster count → Fit K-Means: sets K
- Prepare features → Fit K-Means: provides features
- Fit K-Means → Visualize clusters: projects clusters
- Fit K-Means → Summarize user profiles: summarizes groups
- Summarize user profiles → Recommendation and UX proposals: motivates proposals