Goddag! 🇩🇰

This week, we continue riding the wave of computational tools being used in literature 📊📚 (check out ⚡️Trendbreak #21⚡️ for a recap). A few years ago, The Atlantic published a riveting piece on harnessing ML and data analysis to quantify and highlight what many authors already knew about "narrative shapes" in storytelling. The article draws on American novelist Kurt Vonnegut's fierce advocacy of this theory, and the findings of a study that put it to the test with over 1,300 texts from Project Gutenberg (a collection of over 60,000 public domain e-books).

We've discussed Python's 🐍 linters (auto-formatting tools) before, but did you know there are similar tools for SQL - often ignored when it comes to stylistic standardization? Meet SQLfluff, a relatively new open-source project that works wonders whether used in command line, as a pre-commit hook, or as a VSCode plugin. It's well-explained in this blog post.

Ever heard of online learning? It's a unique scenario where all the data isn't available at the learning stage, or it changes rapidly over time, demanding models to be learned and updated on-the-go. It's a no-brainer for applications like cybersecurity (fraud detection or intrusion attempts) 🥷 or large-scale recommendation systems.
Max Halford offers a great primer in this presentation at the GAIA 2022 conference, introducing RiverML, his brainchild and open-source project that rounds up tools with an API similar to scikit-learn.

Happy reading and have a great week! 😀