Skip navigation

Boulder

3 Posts authored by: Steven Rouk

The competition for Iron Viz qualification has begun! In case any of you don't know about the Iron Viz, it's an "Iron Chef"-like competition that Tableau holds every year at their annual conference. Here's a blog post that talks a little more about it: Iron Viz Competition 2015.

 

Everyone has submitted their visualizations for this first round using only data from Wikipedia, and there are a lot of good ones out there to learn from. I know that I'll personally be downloading several of them to learn from, maybe you'll find some things you want to learn about too!

 

Iron Viz Wiki Data Submissions

 

For my submission, I chose to manually scrape the "Notable Deaths" data from Wikipedia, a task which ended up involving way more data cleaning than I thought it would be. The data resided in 29 individual pages, links to which can be found in this one central page. Unfortunately, although the data was roughly in the same format from page to page, it differed just enough to make it difficult to structure into a Tableau-friendly format. The process I went through was a long combination of manual work and Python scripting, in order to turn the data into a consistently-structured .csv file that could then be loaded into Tableau. (Headers: Date of Death, Name, Description, Birth Year, Age)

 

After finally getting the data into Tableau, I decided to try to parse out the Nationality, Occupation, and Cause of Death for each person using a long series of IF / ELSEIF statements. (I also used CASE / WHEN statements for a few applications. I prefer to use these since they're more compact, but they couldn't do the string parsing that I wanted.) Not a perfect method because of complexities within the data, but it ended up returning some very interesting results nonetheless! Check out the "About" tab in the viz for more info on the methodology. (The viz is below, but here's the Tableau Public link if you want.)

 

http://public.tableausoftware.com/static/images/Wi/WikiR_I_P_/NotableDeaths/1.png

 

By the way, if you aren't already active in the Tableau Public world, I definitely recommend finding and following some bloggers! I recently had the opportunity to talk with Jewel Loree, and she is always coming out with plenty of interesting vizzes that will improve you analytically and creatively. (VizCandy and Paint By Numbers are a couple others I try to keep up with, as well.)

 

Happy exploring!

 

Steven Rouk

steven@boulderinsight.com

A few weeks ago, a security researcher released a dataset of 10 million usernames and passwords that he'd scraped from places online. (He cleaned them up a bit, and they're probably all dead, so don't worry.) Since a release of that magnitude hadn't really been done before, I thought it would be really interesting to analyze the data in Tableau and see what insights I found.

 

The results? Lots of weak passwords, a smattering of profanity, and even some Bieber references.

 

The full blog post goes into more detail, including how to use Tableau 9.0's new support for regular expressions (which are completely and totally awesome). Here's the link – 10 Million Passwords Analysisor just click the image below. Make sure to explore the viz for all the cool insights! (Tab #3 will let you search a subset of usernames and passwords for any words you want.)

 

blog - viz screenshot smaller.bmp

Steven Rouk

MetaViz of the Day

Posted by Steven Rouk Mar 2, 2015

Hello everyone in the Boulder Tableau User Group!

 

This content page is looking pretty dead, so I thought I'd write a post to add an infinitesimal amount of life here. (Would that make the page "undead"? Yikes, zombie page!)

 

We lovely folks here at Boulder Insight like to keep up with the "Viz of the Day" publications from Tableau Public, and we thought it might be Viz-of-the-Day-worthy to analyze all of Viz of the Days! (Vizzes of the Day? Neither sounds good.)

 

So, without further adieu... the MetaViz of the Day!

 

Some interesting things to look for:

  1. Sports holds the title of most-talked-about-topic, and it's also the category that has seen the greatest increases in the past couple of years.
  2. John Schoen and Ramon Martinez have a whopping 35 vizzes between them. If you want to get on Viz of the Day, apparently they're the ones to learn from!
  3. The number of Economy vizzes published per year has decreased the most consistently out of the categories. Perhaps this is because the economy is continually doing better from year to year? Or maybe people have lost all hope and choose to watch football in despair?
  4. The number of vizzes about Culture spiked in 2013. Why was 2013 such a hot year for cultural topics?

 

 

Feel free to let us know of any other insights you find! And maybe one day, your viz will be listed here too...

 

Happy Visualizing!

 

Steven Rouk

Visualization Scientist

Boulder Insight