Perl Monk, Perl Meditation | |
PerlMonks |
comment on |
( [id://3333]=superdoc: print w/replies, xml ) | Need Help?? |
Well BrowserUk, i just finished some testing of my approach. I was able to search a dictionary of 30 million words for one of 1245 possible fuzzy matches in 30 minutes on 555 mhz pc. Admittedly this was anchored at the first char not a match "in string" as it were. Considering your extremely harsh comments in the CB id be very happy to run whatever code you choose against my code to see which is faster. The code takes as an argument a string of digits and then finds all 0 1 or 2 char mismatches of that string in the dictionary with the correct number of digits. If you provide code so that I can do the equivelent of
Then ill run your code against mine with the same test set, etc and we will see whos solution is faster. Since my solution represents a special case of the general problem yours should perform even better than you have suggested here so I expect that youll win. But until you have dont you think you should keep your comments at least polite?
--- demerphq In reply to Re^2: Fuzzy Searching: Optimizing Algorithm Selection
by demerphq
|
|