I’ve been lucky this semester. Although I’m enrolled in three courses, all the work in those courses rolls into my thesis somehow. In essence, I’ve managed to talk my way into having just one very large assignment due at the end of April. The upside is that I get to expend my energy into something I care about and get a jump start on my Ph.D. research. The downside is that it’s a lot of work that I have to, you know… do.
When I say I’m studying “computational linguistics,” most people say “what’s that?” I say don’t worry about it: it’s only the stuff that makes Google and Siri work, and much more besides. I work in the “much more besides” area—specifically semantics. I’ll maybe get into more details later, but suffice it to day that I research why things mean what they mean, and it does’t get much more esoteric than that.
I’m approaching the phase that every computational linguistic project enters at some point: that of throwing gobs of data at the problem until it gets better. I get to wade through reams and reams of text looking for verbs, pulling out patterns, and trying to figure out how to mark it all up so I can feed it all to a computer program at some later date. If that isn’t fun, I don’t know what is. Before I get there, however, I get to nail down the theoretical background of the work. Given that much of this amounts to “what makes a word that word?” this is a bit like trying to wrestle a balloon animal underwater.
PhDComics has a good one about how this feels, though my advisor is a lot nicer about it.
