ISSUE 362  ·   November 16, 2021

Trends

Data Librarian: Next Generation Of Data Engineering

Supporting an experimental data science process requires the role of data engineering to evolve. At the least, data engineers need to be data science literate because curating data requires an understanding of what it takes for data to be useful. The role described here is much more than building pipelines and collecting metadata...
Vin Vashishta

Sponsored Link

Apply today: Master of Data Science for Public Policy

Passionate about data and its role in public affairs? Join the Master of Data Science for Public Policy (MDS) at the Hertie School in Berlin, Germany! This two-year master’s taught entirely in English welcomes students from all academic disciplines who want to use data to solve real-world challenges. Explore the MDS and tune in to our Open Week from 6-10 December to learn more.

Tutorials, Projects & Opinions

Building confidence in a decision

Decision making under uncertainty can be difficult and with A/B testing, results aren't necessarily a reflection of the underlying truth. Here's how Netflix looks beyond just the numbers to decide if the results of an A/B test are compelling.
Netflix Technology Blog

Why machine learning hates vegetables 

In this personal account of automated "intelligence" going awry, Emily Riederer explores why an ML product would recommend that she stop eating vegetables. Ultimately, it's a cautionary tale of what can happen when out-of-the-box algorithms are applied to real-world problems. 
Emily Riederer

Root Causing Data Failures

Finding the root cause of a data quality issue can take a lot of time and once you understand where a problem is, many things need to happen to resolve the issue and get the data back in good shape. But what if you could find and resolve issues automatically? 
Anomalo | Vicky Andonova

How to think about the ROI of data work

Great post that explores how each role on a data team can have the highest possible impact. It turns out that Graph Theory is a surprisingly useful concept for estimating the ROI of data work.
Mikkel Dengsøe

How to Get Strategic Value from Data Analytics

A significant proportion of organizations are neither making the
most of their data nor fully leveraging the power of their analytics.
Infusing analytics into workflows and embedding analytics into
products provide a solution to this problem. Learn more.
// sponsored

Code & Tools

doubtlab

Doubt your data, find bad labels. This repository contains general tricks that may help you find bad, or noisy, labels in your dataset. 
GitHub | koaning

Data Visualization

Scientific Visualization: Python + Matplotlib

Awesome new open-access book on scientific visualization using Python and Matplotlib. Covers Matplotlib fundamentals, visualization design, advanced topics such as optimization & animation, and concludes with a showcase of examples. Free to download.
Nicolas P. Rougier

Clarity and Aesthetics in Data Visualization

Great set of guidelines to help make your visualizations visually appealing and easy to understand. Includes lots of examples.
FILWD | Enrico Bertini with Daniel Kerrigan 

How did you like this issue of Data Elixir?

👎    1     2     3     4     5   👍

Was this email forwarded to you?  Sign up here >>