AminuAhmadh/naija-sentiment

# 🇳🇬 Naija Sentiment v2 > Real-time Twitter sentiment analysis for Nigerian political parties — APC & ADC Built with VADER + Nigerian political lexicon, Google Gemini AI summaries, and a dark-mode Streamlit dashboard. --- ## What's new vs v1 | Feature | v1 | v2 | |---|---|---| | Sentiment engine | TextBlob (movie review corpus) | VADER + Nigerian slang lexicon | | Nigerian language support | ❌ | ✅ Pidgin, political slang, key terms | | Dashboard design | Basic | Dark mode, Plotly charts, animations | | AI summaries | Gemini | Gemini (improved prompt) | | Demo mode | ❌ | ✅ Works without any API keys | | Comparison tab | ❌ | ✅ Head-to-head APC vs ADC | | Score distribution | ❌ | ✅ Histogram + timeline | | Tweet feed | Basic | Filterable, sorted by score | --- ## Quickstart ```bash # 1. Clone git clone github.com cd naija-sentiment # 2. Install dependencies pip install -r requirements.txt # 3. Set up environment variables cp .env.example .env # Edit .env with your API keys # 4. Run the dashboard streamlit run app.py ``` ### No API keys? No problem. The dashboard has a built-in demo mode with realistic mock tweets. In the sidebar → Controls → **Use demo data** → **Load Demo Data**. --- ## API Keys ### Twitter Bearer Token - Go to developer.twitter.com - Create a project and app (free tier = Basic, ~$0/month for limited access) - Copy your **Bearer Token** into `.env` ### Google Gemini API Key - Go to ai.google.dev - Free tier available, no credit card needed - Copy your key into `.env` --- ## Project Structure ``` naija-sentiment/ ├── app.py # Main Streamlit dashboard ├── src/ │ ├── scraper.py # Twitter API v2 fetcher (Tweepy) │ ├── sentiment.py # VADER + Nigerian lexicon analyzer │ ├── summarizer.py # Gemini AI summarizer │ └── mock_data.py # Demo data generator ├── data/ │ ├── apc_tweets.json # Fetched/cached APC tweets │ └── adc_tweets.jso …