# Worship Engagement Prediction System for Nigerian Gospel Artists
A machine learning-powered system that predicts worship engagement metrics for Nigerian gospel artists using Spotify data. This project aims to help artists, labels, and industry stakeholders understand and predict audience engagement patterns in the Nigerian gospel music market.
## 🎯 Project Overview
This system collects and analyzes Spotify data from Nigerian gospel artists to build predictive models for engagement metrics such as:
- Artist popularity trends
- Follower growth patterns
- Track performance indicators
- Weekly engagement metrics
## 📊 Current Progress
### ✅ Completed Features
#### **Data Infrastructure**
- **Spotify API Integration**: Complete authentication and data fetching system
- OAuth2 client credentials flow
- Artist profile data retrieval
- Top playlist tracks fetching (Nigeria market)
- **Data Ingestion Pipeline**: Automated data collection system
- Batch processing for multiple artists
- Data transformation and cleaning utilities
- Error handling and logging
- **Database Schema**: SQLite database with three main tables:
- `artists`: Artist profiles (ID, name, genres, source)
- `tracks`: Track metadata (ID, artist, name, duration, release date)
- `artist_weekly_metrics`: Time-series metrics (followers, popularity, week tracking)
#### **Core Functionality**
- ✅ Spotify API authentication (`src/data_ingestion/auth.py`)
- ✅ Artist data fetching (`src/data_ingestion/spotify_fetch_artist.py`)
- ✅ Playlist/track data fetching (`src/data_ingestion/spotify_fetch_playlist.py`)
- ✅ Data transformation utilities (`src/data_ingestion/tidy_json.py`)
- ✅ Database operations (`src/storage/`)
- ✅ Comprehensive logging system (`src/config/logger.py`)
- ✅ Batch ingestion script (`src/run_ingession.py`)
#### **Data Collection**
- ✅ Gospel artist ID list (`data/Gospel_artist_ids.csv`) with 60+ Nigerian gospel artists
- ✅ Automated ingestion pipeline for batch processing
- ✅ Error handli …