Reading List

Anecdotally, it would seem that you can be on Twitter, or you can attempt to find joy, health, and balance in your life, but not both. The writers, journalists, and niche food-opinion-havers in my timeline form a fairly diverse lineup. What unites us is a strong ambivalence—yes, I’m mincing words here—about the space itself: the “hellsite” we can’t leave behind.

Digging into the TDS archive these past few weeks has had the unexpected effect of suggesting a different possibility—an alternate reality, even. Here were dozens of data scientists and AI experts spending massive amounts of time on Twitter and… doing productive things with it?! Not falling into ever-wilder spirals of despair?! Drawing thoughtful insights about the world?!?!

How was any of that possible?

It’s clear to me now that my use of the platform as a primary source of news has dramatically shaped my perception of it. If you mostly go on to Twitter to stay abreast of [waves hands randomly and indiscriminately at the world], it would stand to reason that the emotions you have about the news eventually converge with the ones you feel about the app. On the flip side, taking a step back to analyze how other people and communities use the platform — which Twitter has made possible thanks to its robust API—lends itself to blissful detachment (or at least a semblance of it), one that I one day hope to achieve as well.

To prove my point, here are some of my favorite TDS Twitter/data science crossover posts — read them! All of them, some of them, a couple; you won’t regret it. The archive runs extremely deep, so I assembled this collection after some major filtering, and across three categories: new and noteworthy, hands-on resources, and all-time greats. Let’s dive in.

New and noteworthy

Did I mention I use Twitter mostly for news consumption? The reason is the platform’s undeniable ability to capture political and cultural moments — sometimes it is the moment. From Black Lives Matter (and the occasionally fraught ways companies have attached themselves to the social movement), to Twitter users’ jubilant reception of Netflix hit Bridgerton over the holidays, these recent posts apply the tools of data science to study such moments in depth, and do it extremely well.

Hands-on resources

Many of our readers come to TDS because our community is where they find answers to the practical challenges they face in their work, in their studies, or in their passion projects. Twitter data analysis is no exception, and the posts gathered here provide clarity and step-by-step instructions on how to collect, clean, process, and draw insights from tweets.

All-time greats

Yes, our archive on this topic is immense, but some posts — unlike most tweets!—have really stood the test of time, and are just as sharp and engaging today as they were when their authors first published them. They cover a wide range of topics, from Twitter’s own hiring process to detecting signs of depression in tweets’ linguistic markers. But first: CUTE DOGS.

Let us know if any of the above has resonated with you in particular. Also: have you read a great post—here or on another site—featuring a Twitter-related data project? Have you written one yourself? Share it with us in the comments. Also-also: are there other topics you’d like to see us cover in a future reading list? Tell us that, too.