data, data
Data, data is a deep dive into the sonic and lyrical universe of Jorge Drexler — exploring his work through the lens of data science and music theory.
I analyzed all of Drexler’s officially released songs using APIs, scraped lyrics, and combined tools from natural language processing, emotion modeling, and music theory. The result? A multi-layered view of the artist’s evolution, themes, and creative patterns.
- For data collection i used pandas, BeautifulSoup, Spotipy, the Genius API, and Spotify's Web API
- Analysis and modeling using pandas, NumPy, Matplotlib, Seaborn, scikit-learn, SciPy, NLTK, wordcloud, and py-lex
- All wrapped in Python 3, Jupyter Notebook, and the occasional PyCharm pass
Beyond the data, what made this special was that Jorge Drexler himself acknowledged the work — in a tweet that completely made my year.
visuals
🧠 Emotional trends in Drexler's lyrics over time
🎵 Tempo patterns by album — from slow ballads to faster experiments
🎼 Most common keys used across his discography
📝 The songs with the most lyrical density
☁️ Wordcloud of Drexler’s most frequent words
📚 Comparing lyrical and lexical complexity
🧩 Correlation matrix between emotional markers
from the artist himself
press
- El Observador – one of Uruguay's major newspapers (🇺🇾 Spanish)
- Redacción – Argentinean digital media spotlight (🇦🇷 Spanish)