— In the News —
The American Statistician released a special issue this month that calls for an end to the entire concept of statistical significance. This article in Nature offers a nice overview of the key points and got a lot of attention around the web this week.
— Profiles —
Demis Hassabis founded DeepMind with the mission to "solve intelligence." From there, the plan was to use that intelligence to solve everything else. Ambition like that is a long game and along the way, Google bought DeepMind which gave Hassabis access to vast resources. This article in the Economist explores the mission, the realities, and where it's all going.
— Sponsored Link —
Realize the promise of data analytics and find the opportunity in the numbers. The Master of Management Analytics from Smith School of Business is essential training to unleash the potential of data and generate competitive advantage.
— Tools and Techniques —
Great code-through for building free, privacy-focused analytics for websites and applications. Covers requirements, tech considerations, data handling, analytics and rationale for decisions made along the way.
This series of short videos make linear regression visually intuitive and are great explainers.
Qri is a version control system for datasets; built on top of the distributed web. It's kind of like Git for data and it allows datasets to be easily discovered and kept up-to-date. The discussion on Hacker News about this project is also worthwhile and offers some ideas for other approaches.
This post on the TensorFlow blog introduces a new library in TensorFlow Probability for forecasting time series using structural time series models.
In part 2 of a five part series about productionizing data science, Schaun Wheeler
explores portability as a strategy for leveraging your code when your goals change.
This articles explores why differential privacy is one of the strongest de-identification privacy enhancing technologies, how it mitigates re-identification risks while deriving insights and utility from data, and why it needs to be top of mind for data science teams.
— Resources —
In this post, Tomislav Hengl explores the critical role that data science plays in monitoring Earth's environment. Includes numerous links to datasets, projects and research.
— Data Viz —
Plotly Express is a new open-source Python visualization library that exposes a simple syntax for complex charts. It’s a wrapper for Plotly.py and was specifically designed to have a terse, consistent and easy-to-learn API. This is a nice introduction to the project, including lots of screenshots and a Jupyter notebook of examples to get started with.