Creating a faceted classification system for cat names#

Thistle does not know his name. He occasionally comes when we make kissing noises (haven’t tested if this is a greater-than-chance occurrence). But he’s so smart! Why doesn’t he know his name? Spoiler: it’s our fault.

I’m in a course called Foundations of Resource Description and Knowledge Organization. A mouthful. But incredibly fun. You may not believe me if you do not know that I’m the kind of person who, when looking up cozy video games to try, gets excited about the video game where you get to take stuff out of boxes and put it on shelves. My favorite job ever was organizing the supply room at an elementary school. Picture this. Paper mayhem. Different colors, weights, sizes, all mixed together, everywhere. Agony. But I BROUGHT ORDER TO CHAOS *MANIACAL LAUGH* (silent maniacal laugh, of course; I struggle to laugh out loud, though sometimes I manage to make guinea pig noises when I’m laughing hard… Maybe picture an evil guinea pig laughing). I made categories. I made labels. I realized I made more categories than cubbyholes. I remade the categories and labels. It was blissful. I am drooling and digressing.

The assignment#

For this assignment, we had to make a faceted classification system for a set of items of our choosing. I initially chose data visualizations, a special interest of mine. I stayed up way too late, my brain exploded, and Colin jokingly suggested I pick something simpler, like names we call our cat. Fine, I will be silly and maybe explode less and be able to sleep. The latter did not happen. I hated and loved it.

This kind of system is supposed to be precise enough that different people would end up at the same classifications if they followed the class hierarchy and descriptions. So I figured, if anyone should be able to do it, it should be automate-able, right? right? wrong. Or at least, I tried and it didn’t work and I wasn’t willing to put in more effort. I thought making the figure would help me figure out how to make the package, but it is surprisingly hard to automate classifying a name like “Honey” as being used when Thistle is acting cute. Any cat lover who hears me call Thistle “Murder muffin” should be able to guess his energy level, at least. Maybe GPT could do it. But it was too hard for SpaCy’s sentiment analysis, which I thought would be a good proxy.

My faceted classification system#

Use Case. This system classifies names that my husband and I call our cat. We have just over 30 names for him. I have chosen to limit names to being one or two words long because that includes nearly all the names. That decision also kept the task manageable.

As an extra challenge, I created a data visualization of the scheme. As someone who has used text-mining methods in research, I am interested in the problem of how to visualize text data. In addition to using the spatial layout / axes (for the “Properness” and “Realness” facets), I used the text properties of font type (for “Poesy”) and font size (for “Frequency”), as well as highlight color (for “Arousal”).

Commentary#