From the course: Data Science Foundations: Data Assessment for Predictive Modeling

Unlock the full course today

Join today to access over 22,600 courses taught by industry experts or purchase this course individually.

A strategy for processing data sources

A strategy for processing data sources

From the course: Data Science Foundations: Data Assessment for Predictive Modeling

Start my 1-month free trial

A strategy for processing data sources

- [Male Speaker] I want to share a strategy and a metaphor to help you with how to go about choosing data sources. I'm going to give you a suggestion about where and how to look for data. What many are tempted to try, at least at first, is to start small with just one or two data sources. They start by asking a subject matter expert for advice on where to look first, and then they work with one major data source at a time, lets say, social media, then they rush right to modeling. The idea is that it will be more efficient. This is especially tempting if you're working in sprints and you're feeling time pressure. I want to explain the risks of that approach using our metaphor. Perhaps not the metaphor you were expecting. It works like a dog show. I'll explain the dog show metaphor in just a moment, but first, a bit of theory. The reason that folks want to start slow and fail fast, is that they're afraid that the data…

Contents