SF Data Weekly - Data Smellification, Clean Lakes and Redis

Alright, this one is just for fun: using smell as an alert when browsing unsecured websites over unsecured wifi, data sonification apps, touch and play-doh visualizations.

Some good tips on when to use – and when not to use – Lambda to process Kinesis data streams.

Register for Xplenty’s live demo this Thursday @10am PST! You will learn how to build data pipelines that help organizations cleanse and mask their sensitive PII data, before centralizing it in a database. [Sponsored Content]

Athena now has CTAS (CREATE TABLE AS SELECT) support, which is a great convenience and also gives you the ability to leave a lot of temporary tables around. This piece shows an automated way to clean up after your CTAS-using processes.

Uncovering what the capabilities of the world’s most popular in-memory data store are and how to leverage its potential.

Understanding population mobility patterns is important to policy makers, marketers, business owners and now in particular, epidemiologists. This piece uses a Tweepy stream listener to grab the data, and then picks the geotagged tweets to look at mobility in New York and parts of New Jersey bordering New York.

Part of a series of articles analyzing and visualizing COVID data. Spoiler alert: the risks are for the old, not the young.

Some great code examples to generate sophisticated graphs from a biostats dataset.

AirBNB engineers share their definition of a gold standard, called “Midas” internally, and how data is certified to that standard. Part 1 of the series is worth a read, too.

Five tips from Carl Gold, Chief Data Scientist at Zuora, to help combat model drift, where the training dataset no longer represents the external world.