>>You can use my new Similarity Algorithm based on WikipediA (SAWA):
>>http://tinyurl.com/sawa-filannino
>>
>>Try it and say me what do you think!
>
>I'd love to try it but got 404'd.
You can also use this URL:
http://193.204.187.223:8080/sawa/
~~
How about:
http://www.cflib.org/udf/levDistance
On Sat, Oct 2, 2010 at 9:46 AM, Larry Lyons wrote:
>
>>You can use my new Similarity Algorithm based on WikipediA (SAWA):
>>http://tinyurl.com/sawa-filannino
>>
>>Try it and say me what do you think!
>
> I'd love to try it but got 404'd.
>
>
~~
>You can use my new Similarity Algorithm based on WikipediA (SAWA):
>http://tinyurl.com/sawa-filannino
>
>Try it and say me what do you think!
I'd love to try it but got 404'd.
~|
Order the Adobe Coldfusion Anthology now!
http:
> I'm looking for an algorithm that will take two pieces of text as
> input and return some comparison of those texts, and how similar they
> are to one another. Ideally I would like it to return a number which
> would indicate the level of similarity, but I can probably work with
> just abou
>Post your email and will contact you directly.
My email is chrisl [symbol] higheredjobs [you know] com.
~|
Create Web Applications With ColdFusion MX7 & Flex 2.
Build powerful, scalable RIAs. Free Trial
http://www.adobe.com/pro
Post your email and will contact you directly.
> Thank you so much for the link! That was exactly what I was looking
> for.
>
> > Let me know, if you need a CFX tag for this ($$).
>
> Since I've never tried creating my own CFX tag before, I'd be
> interested. How much are you
asking?
~
Thank you so much for the link! That was exactly what I was looking for.
> Let me know, if you need a CFX tag for this ($$).
Since I've never tried creating my own CFX tag before, I'd be interested. How
much are you asking?
~
> I know that this problem has been solved as an algorithm, because the
> web searches I've done have indicated that. However, the only two
> links to the actual algorithms that I've found are both broken.
Yes, these algorithms calculate "Editing distance" and well known since late
60s. Most
>When a user changes a display, have the old and new text emailed to an
>address set up for the purpose. This will let you see who is making
>lots of changes, and let you check to see if any of those people are
>making major adjustments.
Please don't take offense to this, but that's a pretty p
It won't automatically catch people, but
When a user changes a display, have the old and new text emailed to an
address set up for the purpose. This will let you see who is making
lots of changes, and let you check to see if any of those people are
making major adjustments.
Admittedly, i
> What sort of text?
The text is about one to three pages in length (in general), and we charge per
entry. I'm trying to allow editing so that the customer can make minor changes
or updates, but not large-scale editing in an attempt to circumvent charges.
This is definitely a large scale syste
What sort of text?
If it's long written stuff, I'd split it into paragraphs then loop through each
paragraph and split into sentences, then loop through each sentence and split
into words - count the number of matches of words in the related sentence in
the original text and if a sentence a sco
I have used an md5() hash for this but we were only looking to see if there
was change not the amount of change.
Just throwing out ideas :)
--
~Eric
~|
ColdFusion MX7 and Flex 2
Build sales & marketing dashboard RIAâs for yo
> "Similar" is a rather vague description. Are you looking for overlap?
> Length? Number of characters in the same location in each string?
> Number of similar characters?
Good point, pardon my lack of clarity. I'm looking for matching text. It
would essentially be identical to a plaguarism
Are you talking about matching sections of text (copyright/plagerism issues)
Rough similarity (similar size, number of words)
Similar content groupings (word count, number of matches)
Whole phrase matching
On 3/5/07, Ben Doom <[EMAIL PROTECTED]> wrote:
> "Similar" is a rather vague description.
"Similar" is a rather vague description. Are you looking for overlap?
Length? Number of characters in the same location in each string?
Number of similar characters?
Soundex is a good start if what your are looking for is similarity in a
sounds-like or spell-check sense, but isn't very helpfu
Try Soundex (http://en.wikipedia.org/wiki/Soundex). Its for names but may
get you started in the right direction.
~Max
On 3/5/07, Chris Long <[EMAIL PROTECTED]> wrote:
>
> I'm looking for an algorithm that will take two pieces of text as input
> and return some comparison of those texts, and how
I'm looking for an algorithm that will take two pieces of text as input and
return some comparison of those texts, and how similar they are to one another.
Ideally I would like it to return a number which would indicate the level of
similarity, but I can probably work with just about anything.
18 matches
Mail list logo