I need a sabbatical!

[robinturner]

I could really do with a year off to have a good look at Latent Semantic Analysis. I mean, even the name is cool! Call me a sad geek, but this kind of thing gives me a thrill:

Latent Semantic Analysis (LSA) is a theory and method for extracting and representing the contextual-usage meaning of words by statistical computations applied to a large corpus of text. The underlying idea is that the totality of information about all the word contexts in which a given word does and does not appear provides a set of mutual constraints that largely determines the similarity of meaning of words and set of words to each other. The adequacy of LSA's reflection of human knowledge has been established in a variety of ways. For example, its scores overlap those of humans on standard vocabulary and subject matter tests, it mimics human word sorting and category judgments, simulates word-word and passage-word lexical priming data and, as reported in Group Papers, accurately estimates passage coherence, learnability of passages by individual students and the quality and quantity of knowledge contained in an essay.

There again, if I really did get a year off, I'd probably bugger off and spend it learning t'ai chi or capoeira, returning with nothing to show for my sabbatical but an improved ability to beat people up. I'm perverse like that.

Comments

[b0rg.livejournal.com]

I'd say t'ai chi, or even a flute somewhere in the mountain monastery sound much better :)

[philipispdr.livejournal.com]

Fascinating stuff, thanks for linking.

[trochee.livejournal.com]

I think LSA is pretty cool. I did a bunch of work on it in my previous life, and it's a very interesting technique. Kind of heavy on the math, though -- let me know if you want a reading-list or pointers.

Hmmm...

[ankh156.livejournal.com]

Seems to be a 'statisticalised' variation on the "meaning = use" hypothesis (LW).

[wabi.livejournal.com]

Thanks for that link. I've tried to understand LSA before, but always end up getting hung up on the math. Who can really imagine 1000-dimensional space?