Re: Near Text Search

Devin Asay wrote: > Look in User Samples (revOnline) for my "Spelling Matching with > Soundex" stack. I think I classified it with tags "text" and > "algorithm". It's just a modification of the basic SoundEx algorithm > with a couple of my own modifications thrown in. Found it - thanks! -- Ric

Re: Near Text Search

BobS wrote: > That is kind of what I was doing, ... Does that mean I’m a genius? Instead of Levenshtein Distance, I think we need a Euclidean Distance formula to calculate that. ;) Whenever I run it on myself the result always approaches zero. JimL ___

Re: Near Text Search

Look in User Samples (revOnline) for my "Spelling Matching with Soundex" stack. I think I classified it with tags "text" and "algorithm". It's just a modification of the basic SoundEx algorithm with a couple of my own modifications thrown in. Devin On Feb 26, 2014, at 8:38 PM, Richard Gaskin

Re: Near Text Search

Oh right! Duh… On Feb 26, 2014, at 20:54 , Peter Haworth wrote: > You don't need another column. > > If you don't get an exact match > > SELECT soundex(name)=soundex("Wittier Fun Center") > > Not sure of performance implications. > > Pete > lcSQL Software > On Feb 26, 2014 8:12 PM, "Bob Snei

Re: Near Text Search

You don't need another column. If you don't get an exact match SELECT soundex(name)=soundex("Wittier Fun Center") Not sure of performance implications. Pete lcSQL Software On Feb 26, 2014 8:12 PM, "Bob Sneidar" wrote: > I just did a query on soundex for 3 values: Whittier Fun Center, Whitier

Re: Near Text Search

That is kind of what I was doing, only a little different.I was doing a kind of binary bit significance on each character and totaling the ascii values times the bit significance. Does that mean I’m a genius?? no… that can’t be right… Bob On Feb 26, 2014, at 20:04 , Jim Lambert wrote: > Rich

Re: Near Text Search

I just did a query on soundex for 3 values: Whittier Fun Center, Whitier Fun Center and Wittier Fun Center. All returned the same result. I think what I will do is create a column populated with the soundex of the customer name as a secondary lookup when I don’t get an exact match, then alert th

Re: Near Text Search

RichardG wrote: > > I'm still missing it - where can one find Devin's Levenstein function? Try Rishi's http://www.rghosh.com/code/levenshtein-distance-livecod/ And yes, that ‘livecod' is correct, though a bit fishy. Jim Lambert ___ use-livecode mai

Re: Near Text Search

I'm still missing it - where can one find Devin's Levenstein function? Bob Sneidar: Don’t know how I missed that, thanks. On Feb 26, 2014, at 18:50 , Peter Haworth wrote: If I remember correctly, Devin has some functions that implement a Levenstein Distance algorithm which will identify pot

Re: Near Text Search

Don’t know how I missed that, thanks. Bob On Feb 26, 2014, at 18:50 , Peter Haworth wrote: > Does mySQL have a Soundex function? > > If I remember correctly, Devin has some functions that implement a > Levenstein Distance algorithm which will identify potential misspellings of > a word > >

Re: Near Text Search

Does mySQL have a Soundex function? If I remember correctly, Devin has some functions that implement a Levenstein Distance algorithm which will identify potential misspellings of a word Pete lcSQL Software Home of lcStackBrowser an

Re: Near Text Search

Hi If I get what you want, why not just append the ASCII values to create a much longer number?: on mouseup answer textToNum("abcdefgh") end mouseup function textToNum theString put lower(theString) repeat for each char theAscii in theString put charToNum(theAscii) after theAsc