Yes, it sounds like something from the Simpsons, but it's a real word used to describe the problem of deriving semantic meaning from lexical cues. Of course, if you're a computational linguist, you already know this. But if you aren't, and this still seems interesting, check out the Making Sense of Sense: Bringing Computation Linguistics and Psycholinguistics Together workshop at this 11th Annual EACL Conference.

Why would this matter to political scientists? Because scholars are beginning to discover that the tools of quantitative linguistics allow us to identify and trace meaning at speeds and scales unimaginable just a decade ago. It's not just the NSA and CIA - check out:

Benoit, K. and M. Laver (2002). Extracting Policy Positions from Political Texts Using Phrases as Data: A Research Note. Midwest Political Science Association, Chicago, IL.

Benoit, K. and M. Laver (2003). "Estimating Irish Party Positions Using Computer Wordscoring: The 2002 Elections." Irish Political Studies 17(2).

McIntosh, W., M. Evans, et al. (2004). Only Words, or Data? Assessing the Relative Policy Positions in Supreme Court Briefs and Opinions. College Park, MD: 50.

Tags: content analysis, text mining, conference

No comments: