— Insight —
Not long ago, Business Intelligence and Data Science were very different worlds but that's changing fast. In this interview, Ashley Kramer explores the Sisense platform and how the world of BI is increasingly colliding with data science.
— Tools and Techniques —
This post from Uber Engineering is the latest in a string of deep dives that explore in-house data discovery solutions. Uber's platform is called "Databook" and this is a comprehensive post that covers a variety of topics including motivations for the platform, how it's evolved, architecture, the data model, how discovery works and lots more.
Prediction markets did pretty well in the 2020 election but they're far from perfect. In this post, Arpit Gupta takes a look at the successes and failings of the prediction markets versus the polls in this election cycle.
This collection of hands-on tutorials walks through a wide variety of machine learning projects using Python.
Great post that explores the various goals, skills, and deliverables of common "data science" job roles and what these evolving titles mean for the industry and your team.
Pluto.jl was built from the ground up to be a fresh new notebook system, written in Julia. Pluto notebooks are stored as executable .jl files and they're reactive, which means that affected cells automatically update when functions or variables are changed.
Dagli is a new open source ML library for JVM languages, like Java. It contains a comprehensive collection of statistical models and feature transformers that are ready to use out of the box. Dagli is intended to make it easy to write, revise and deploy production-ready models.
— Projects —
How effective are stay-at-home orders on slowing the spread of COVID-19? This new project uses smartphone data to model the mobility of 98 million people to track the spread of COVID-19, detect hotspots, and guide reopening strategies. Includes an interactive simulation, links to anonymized data and a paper preview.