Doors open at 6:30
Presentation starts at 7:00
Most of the data in the data science realm is data that organizations collect as a matter of fact. Banks have to keep track of your transactions, insurers have to keep track of your premiums and claims, the NSA has to keep track of your likes and re-tweets; if they didn’t, they wouldn’t be doing their jobs. That data can have value, but it can’t answer every question that your business partners might have.
If you have some questions that need answers, but you don’t have all the relevant data that you need, it’s time for design. Designing an experiment is essentially building the data that you need to answer the questions that you have as efficiently as possible. Let’s talk about what it means to be “efficient” in this context, as well as how to accommodate real world constraints in your designs.----