cucumber Posted March 12, 2008 Report Share Posted March 12, 2008 Ir vajadziba validet ievaditos datus latv, angl un kriev valodas tikai uz brutiem. idejas: 1)uztaisit tris masivus ar visu alfabetu un tad parbaudit (tur tas garums mainisies)? 2)preg_match? 3)kads cits variants Link to comment Share on other sites More sharing options...
Vebers Posted March 12, 2008 Report Share Posted March 12, 2008 Ja tev ir lietotaja ievaditais teksts un ir janosaka valoda kada tas ir rakstities (attiecigi lv, en, ru) tad prieks krievu valodas vari megjinat atrast kadus populaarakos kirilica burtus. Prieksh anglju un latviesu valodas uz isiem tekstiem tas ir visai nereali. Jo burti reali daudz maz vienadi un nav garantijas ka lietotaajs tiktiesam izmantos garumzimes / mikstinajumzimes. Bet nu visidealakais variants butu ar dropdown kur var izveleties to attiecigo valodu un pieglabat sessijaa / cookie. Link to comment Share on other sites More sharing options...
Aleksejs Posted March 13, 2008 Report Share Posted March 13, 2008 (edited) Skatoties, cik nopietni vajag :) http://en.wikipedia.org/wiki/Language_recognition_chart http://en.wikipedia.org/wiki/Category:Natu...uage_processing http://www.df.lth.se/~triad/krad/markov.php P.S. Nu ja tikai uz burtiem, tad...Kādā valodā ir teikums: "Es nezinu, ko tu gribi". Principā tiek izmantota arī bigrammu un trigrammu analīze. Edited March 13, 2008 by Aleksejs Link to comment Share on other sites More sharing options...
Recommended Posts