Beer-Style-Brewery-Name-04

Visualizing My Craft Beer Consumption with Circular Dendrograms

As a fan of both craft beer and visualization I thought it would be fun to visualize the array of different craft beers I’ve tried over the last year or so. I’ve been wanting to experiment with some of the different chart types available in Density Design’s RAW data visualization tool and thought a dendrogram or circular dendrogram might work to visualize the hierarchy. For the non-data viz geeks out there a dendrogram is a tree diagram used to illustrate the arrangement of the clusters produced by hierarchical clustering. Dendrograms are often used in computational biology to illustrate the clustering of genes or samples. Source: Wikepedia

For anybody looking to try circular dendrograms in RAW, I found some quirks that I’ll explain below.

Incidentally, for any other craft beer fans out there interested in my reviews of some of these along with photos, you can find them on my craft beer blog CraftBeerPhile.

To start, here’s a small portion of a spreadsheet I use to track my new craft beer consumption.

Beer-Data2

For this visualization, I decided to focus just on three criteria; Style, Brewery, and Name and decided to go with a circular diagram, because, even though the circular construction of it likely makes it more difficult to read than a normal dendrogram, I just like the looks of it better, and this is just for  my own fun and amusement, not serious data viz after all.

Here’s a screen shot of what RAW produced with the hierarchy of Style, Brewery, Name:

RAW-circular-dendrogram

The first quirk I came across is that I had to specify a very large radius for this data set to ensure that the text didn’t overlap on the nodes. The image above was set at a radius of 1600 pixels.

RAW lets you export in several file formats including .svg and png. Downloading a .png file for this dataset however, resulted in something that was essentially useless as you can see below. All the text is in the center, not on the branches.

Beer-Brewery-Style-Name-01

That’s pretty worthless, so I downloaded .svg files of the two hierarchies I wanted to look at; Style, Brewery, Name and Brewery, Style, Name. I opened the .svg files with Adobe Illustrator, saved them as .ai files and increased the text size to make it more legible. This is a quite a bit better than the .png file but there’s still some issue with the text being placed on the nodes on the left side of the diagram, making it difficult to read.

Beer-Style-Brewery-Name-01

Hierarchy: Style – Brewery – Name

Click image to enlarge

Hierarchy: Brewery - Style - Name

Hierarchy: Brewery – Style – Name

Click image to enlarge

Turning off the visibility of all of the nodes and reducing the opacity of the branches helped quite a bit.

Beer-Brewery-Style-Name-03

Click image to enlarge

Beer-Style-Brewery-Name-02

Click image to enlarge

And here’s the same two but with a black background and changing the text color to a more beer-like brown.

Beer-Brewery-Style-Name-04

Click image to enlarge

Beer-Style-Brewery-Name-04

Click image to enlarge

The results probably aren’t very enlightening or interesting to anybody but me and the value of using dendrograms to visualize this is only marginally useful. It does give me a quick insight into the style that I’ve consumed most – IPAs. IPAs are not my preferred craft beer, but it’s the craft beer style that is easiest to find where I live. But, the experience was a success in that I learned some of the quirks of RAW’s implementation of these diagrams. Next step, maybe some more traditional visualization of this craft beer data – after I pour myself a craft beer that is. 😉