|
It would be nice to see a similar study on various other languages (ie old
english, modern english, italian, french, spanish, etc)
****************************** Larry Roux Syracuse University lroux@xxxxxxx ******************************* >>> r_zandbergen@xxxxxxxxx 02/21/03 10:31AM >>> Dear all, --- "Anders, Claus" <Claus.Anders@xxxxxxxxxxxxx> wrote: > Hi all, > after modifying my algorithm (discarding single > letter tokens, looking at > pairs only once and computing a relative count), I'd > like to show my > results: > > o 20% > [ ... snipped for brevity ... ] > > Now we have the char 'o', which makes the difference > between 20% of all > similar (=differing in 1 char only) tokens. To be quite honest, this sort of analysis (which seems quite useful although I would not yet know what to conclude from it) would be best done not in Eva but in a more synthetic alphabet, which does not use multiple symbols for Eva-ch and Eva-sh. It is very easy to convert Eva to Currier or FSG once you've done it once. It is even very easy to make your own analysis alphabet on the fly (both just require the use of BITRANS, and a PC that can still work in DOS). Cheers, Rene __________________________________________________ Do you Yahoo!? Yahoo! Tax Center - forms, calculators, tips, more http://taxes.yahoo.com/ ______________________________________________________________________ To unsubscribe, send mail to majordomo@xxxxxxxxxxx with a body saying: unsubscribe vms-list |