From: "Berger, Daniel" <djberge@...>
Date: 2002-06-07T02:17:35+09:00
Subject: RE: Hashes sensitive to simularity



> -----Original Message-----
> From: Thomas Hurst [mailto:tom.hurst@clara.net]
> Sent: Thursday, June 06, 2002 11:42 AM
> To: ruby-talk@ruby-lang.org
> Subject: Re: Hashes sensitive to simularity
> 
> 
> * Ned Konz (ned@bike-nomad.com) wrote:
> 
> > Perhaps Levenshtein distance?
> > http://www.merriampark.com/ld.htm
> 
> Mmmn, the thing is, I really need a key I can use for a database and a
> hash.  Maybe something like an overlong soundex, or a more generic
> hashing algorithm that produces identical hashes up to a certain
> threshold.
> 
> I suppose I might be able to use a less accurate algorithm and use
> a simularity matching algorithm like Levenshtein distance on 
> the smaller
> set.

Perhaps Napster will open-source their fingerprinting technology now that
they're bankrupt?  Hmmm...maybe not.

A search out on Google revealed precious little.  The Perl soundex modules
all appear to be based on the Knuth algorithm.

That's an interesting alogrithm, though.  Maybe teachers should use it to
see if students are cheating off each other. :) (I used to teach).

Regards,

Dan