Re:What problems have you solved using genetic algorithms/ge...
by abecedarius on Saturday September 11, 2010 @03:32PM
(#1682188)
How'd you measure metricality of text? http://github.com/darius/versecop was my own hack at that using the CMU pronouncing dictionary -- I'd like to come back to it sometime.
Re:What problems have you solved using genetic algorithms/ge...
by endtime on Saturday September 11, 2010 @05:00PM
(#1682347)
Well, we had a solution that used the CMU Pronouncing Dictionary but also had a lot of unknown handling on top of it. Then we used a heavily modified version of Levenshtein distance to measure distance from Shakespearean iambic pentameter, with weighted penalties for the various "common irregularities"...we consulted someone in the English dept. to set the weights reasonably. Our metricality algorithm gave Shakespeare's sonnets 0.85, some random Dickens 0.7, and the NYT 0.4 or so...so we think it worked reasonably well.
Happy to send you the paper if you're interested in more depth than the above.
Re:What problems have you solved using genetic algorithms/ge...
by abecedarius on Saturday September 11, 2010 @07:55PM
(#1682659)
Yeah, unknown words were the biggest problem. I'd love to see your paper -- [email protected] if you don't want to post a link here. Thanks!
I wonder if y'all ever went looking for 'hidden verse' in published prose, too.
Re:What problems have you solved using genetic algorithms/ge...
by mdwrigh2 on Saturday September 11, 2010 @10:18PM
(#1682872)
Would you mind sending the paper my way as well? This sounds fairly interesting. [email protected]
* Generating iterated prisoner's dilemma strategies. It's not that this is an intractable problem...but I wanted to see what would happen in a large population of different strategies. This actually resulted in support for some theoretical results I'd read, which was nice. This was for a class paper.
* The Mona Lisa polygon thing that Roger Alsing came up with - wrote this just to test a new GA lib I'd written to learn Python.
* Optimizing a nano-scale emitter for a new scanning electron microscope - I didn't actually solve this problem, as I had to leave the job where I was working on it before the project finished...but I'd like to think that the project will get somewhere one day.
* "Translating" English prose to English verse; that is, trying to increase the metricality of text while minimizing semantic drift. Very, uh, mixed results (I blame Wordnet!). This was also a class project.
I feel like I might be forgetting something, but that's at least most of it.