Hedos — collective mood index

A hedonic thermometer for Spanish-language internet.

Every hour, RSS feeds from BBC Mundo, El País, and La Vanguardia are scanned. Words are matched against the AFINN-ES lexicon and averaged into a single score between 1 (negative) and 9 (positive).

Live site

Hosted on GitHub Pages: https://[your-username].github.io/hedos

How it works

1. GitHub Actions runs hedos.py every hour
2. The script fetches RSS feeds, scores each article, and saves data.json
3. index.html reads data.json and renders the dashboard

Stack

• Python + feedparser
• AFINN-ES lexicon (2,100+ Spanish words)
• GitHub Actions (free hourly automation)
• GitHub Pages (free hosting)
• Vanilla HTML/CSS/JS

Known limitations

• Small lexicon (2,100 words) — many articles go unscored
• No context awareness — "guerra" scores −2 regardless of meaning
• Sources are Spain/Latin America focused, not global
• Measures linguistic affect, not lived experience

Setup

pip install feedparser requests
curl -L "https://raw.githubusercontent.com/jboscomendoza/rpubs/master/sentimientos_afinn/lexico_afinn.en.es.csv" -o afinn_es.csv
python hedos.py

License

MIT