— In the News —
Machines are creating their own kind of intelligence that humans can't fully understand. In some circles that's a problem but we don't fully understand our own brains either - and yet, we accept what they tell us. This is a fascinating exploration of the nuances of intelligence, aliens, and a kind of science fiction that we're already living in.
It turns out our brains respond to mathematical ideas in ways that are similar to how our brains respond to beauty in things like art, music, and nature. Scientists have actually identified one equation, in particular, that wins the beauty contest.
— Tools and Techniques —
Here's a great overview of the data architectures that drive some of the largest businesses on the Internet. Because their products have massive adoption, the data teams at these organizations are continually pushing the limits and redefining what it means to do analytics at scale. This is the state of the art in April 2017.
This post by Julia Silge shows how to mine a collection of texts using tidy data principles and n-grams. This is very well written and includes useful code snippets along the way.
Sebastian Raschka, author of Python Machine Learning, is working on a new book. This time, the subject is Deep Learning and Sebastian is posting chapters on GitHub as they're written. The newly released "Appendix F" introduces NumPy and offers a glimpse into the breadth and accessibility of the upcoming book.
— Resources —
Here's a collection of well-organized data science cheatsheets for R, Advanced R, Python, Numpy, and Pandas.
— Podcasts —
Startups might feel that they can't compete with big companies because the big guys have all the data but the reality is a lot less clear. This is a great discussion about the many ways that startups and small businesses can, and do, successfully compete with big companies.
— Data Viz —
This online book serves as the documentation for the circlize R package and it may be the best package documentation I've ever seen. It covers use cases for circular visualizations, design considerations, detailed instructions for using the package and lots of examples.
Visualization has been around for a long time and it's evolved over the ages. This is an insightful longread about how data visualization has changed as peoples' ability to understand visualizations have evolved. This is a must-read for people that are interested in the craft of data visualization.