How Data Can Make a Difference in the Real World

Our weekly selection of must-read Editors' Picks and original features

Sep 23, 2021

4 min read

Photo by Yannis Papanastasopoulos on Unsplash

The "science" bit in "data science" might evoke images of antiseptic labs and hushed libraries. In reality, most data scientists navigate complex—if not downright messy—systems, processes, and workplaces. That’s not a bad thing: it also means their work directly affects the world and the people around them, sometimes in profound ways. This week, let’s explore some of our favorite recent posts that focus on that powerful connection.

Reframe the way you approach the profession. Eric J. Daza has built a long career as a healthcare data scientist and Data Science statistician in both academic and industry settings. In a recent conversation with TDS, Eric shares some of the most important lessons he’s learned along the way, including the need to approach data science from a business, rather than a scientific angle.
Maximize data’s impact in the nonprofit sector. For mission-driven and resource-strapped organizations, leveraging the power of data science to inform decisions and brainstorm strategies can make a major difference. Hani Azam and the Delta Analytics team compiled the bits of wisdom they’ve collected through their work with numerous nonprofit orgs, with a goal to help them "break down into small easy-to-do steps best practices to make the most of your data."

Make a difference in your company’s bottom line. One of the most direct ways for data scientists working in industry to add value to their teams is to help prevent customer churn. Luckily, say Daniel Herkert and Tyler Mullenbach, "users prone to churn often leave clues to their disposition in user behavior and customer support chat logs." Daniel and Tyler’s guide goes on to show (in great detail) how data professionals can detect these users and support other teams’ work to keep them happy and engaged.
Speed up your machine learning pipelines. The fanciest models and most cutting-edge algorithms don’t mean much if they end up bogging you and your team down, waiting for hours (or days!) before you can share your insights with other stakeholders. Ben Weber introduces us to real-time feature engineering, and explains how you can build ML pipelines that can respond to prediction requests in milliseconds.
Optimize your data science workflow. Moving beyond Machine Learning models to the full, end-to-end data science life cycle, Adiamaan Keerthi shares actionable advice to help you boost your productivity and efficiency. In the first post of this two-part series, he covers Jupyter Notebook tips and includes numerous handy code snippets to save you even more time.
Take a step back to reflect on human and machine cognition. We couldn’t keep all our theory-loving readers hungry for a whole week—especially when we just published Gadi Singer‘s latest article. The practical stakes of his work on the nature of understanding and its relation to the future of AI might still be years away, but you don’t want to miss out on a thought-provoking read like this one.

If your data science work has helped you better understand—or even just see—the world around you, we hope you consider writing about it and sharing it with our community.

Until the next Variable, TDS Editors