SAS BI: Sample Data for Facial Recognition and UFOs
Oh my goodness … Christmas has come early children! You must have been extra good this year because look what all I have found! A few times before I have discussed where to find SAS sample data sources and even how to create your own sample data.
While learning the SAS Business Intelligence package, it is more fun to work with data that you find interesting so you can search out the trends yourself. It also helps you think through how the user would want to setup the data for Web Report Studio reporting and what stored processes you would create to support the system.
Use the following sources to practice building an information map or even get crazy and code yourself a cube using PROC OLAP. Along the way, see what issues you run into trying to create time series data, join tables by variables, how to deal with duplicate entries, and other issues you only run into when you are actually doing it (aka experience).
Get Your Data Geek On
This tip came from Twitter user One R Tip a Day who recommended the Finding Data on the Internet post. Most of the data sources are in CSV format so easy for you to use SAS Enterprise Guide to grab and turn into a SAS dataset in a simple click (oh! snap!).
One site that really intrigued me was Kaggle, who hosts data competitions. This site is like a party just for data geeks! One competition that looked fun to me was Don’t Get Kicked! sponsored by Carvana (Sell no Evil is their motto). You can win $5K by predicting if a car from auction is a lemon. Come on – I know you can do it. As a bonus I’m willing to split the prize with you based on my excellent moral support.
Facial Recognition and UFO Data?
Here’s a website that has facial recognition data for a homework assignment – although I’m not sure what a BI analyst would do with it. There’s even more facial recognition data at the Face Recognition site. Wow, starting to get a little creepy. Remind me to keep my Halloween mask on while in public.
Speaking of creepy here’s the National UFO Reporting Center data. The data is really interesting. The site has it indexed several ways: (i.e. location and date).
I got the data for North Carolina (where I am) and turns out people in Charlotte reported seeing more UFO than other areas of the state. For the past five years, there were more reports during the month of July. Guess the outer space beings prefer that time to vacation as well.
Someone out there should be able to cross index the facial recognition data with the aliens to figure out which galaxy produces the most attractive ones. I’m biased toward the Milky Way.
By the way – why don’t you leave me some comments about some unusual data sources you know about. Also – please leave your email address with SAS Press, so they can notify you when the Building Business Intelligence with SAS book publishes.