Fred Truman
Design & Code Creative
Learning Bit by Bit – Week 05
March 4, 2011
I think this is week five? Oh man, I’m so confused. Regardless of where we currently are in time and space, we’re beginning to talk about part of speech tagging. Although the examples we’ve been working with in this class are written in Eclipse I’m choosing to work in Python because, as I may or may not have mentioned before, I’m getting comfortable and beginning to establish a nice work flow that I’d like to build upon and improve in a thorough fashion.
I’m only just getting started really, but what I’m attempting to do here as an exercise is visualize a source text’s part of speech twists and turns. I’m not going for anything quite this insightful just yet, but I’d like to approach the level sophistication from Stefanie Posavec’s On the Road visualizations.
As a first step what I did was to create a Python program that takes an input text, in this case 2001: A Space Odyssey, and tags each word’s part of speech. When it either encounters a verb, noun or an adjective it prints either a “1″, “2″, or “3″ respectively out to a text file which I then parse in a Processing sketch to spit out corresponding colored blocks. Starting at the top left and moving right and then down, each blip represents one particular part of speech. Specifically here, I’m using purple for verbs, orange for nouns and green for adjectives. Words that do not belong to one of these parts of speech is represented by a grey dot.
Obviously, this is a very simple exercise to get me up and running. The next step would be to add in a more meaningful selection of five different POS tags (one for each direction in three-dimensional space) and then plot the data. I would like to simplify my workflow by looking into pyglet or some other python compatible graphics engine as well.
Source code to come once I get a proper popup working :-P