Results 1 to 3 of 3
  1. #1
    Join Date
    May 2004
    Moscow, Russia

    Thumbs up Unanswered: Calculation of a degree of similarity of phrases

    i make The Extended stored procedure for MS SQL Server which
    Calculation of a degree of similarity of phrases.

    One of the most complex and important problems for the developer and the operator of a database is maintenance of uniqueness names in the most important references of system.
    Offered function can be used in SQL inquiry as criterion of sorting of the directory according to similarity with a required phrase.

    incredibly high speed of data processing
    Unique algorithm analyzing similarity of phrases even at significant divergences in required phrases
    is not required installation of additional libraries to each client - library DLL must benn installed only on a server.
    Result is all the table sorted in decreasing order phonetic similarity (probably use of operator TOP for sample only the limited quantity of the most similar variants)
    Use of user server function supposes use in Stored procedures, Views and any SQL expressions
    Spends a minimum of server memory

    If the decision of the given problem is interesting to you and there is an opportunity desire and an opportunity to test
    Dll and sample scripts in rar archive (2,5 KB)
    I thank in advance

  2. #2
    Join Date
    Jun 2003
    Provided Answers: 1
    I'd be happy to take a look at your code, but I got nothing but junk from your download.
    If it's not practically useful, then it's practically useless.

    blindman "sqlblindman"

  3. #3
    Join Date
    May 2004
    Moscow, Russia


    It is offered to not analyse a code.
    To estimate an acceptability of the given way for the decision of a specific target:

    Speed of data processing ( length of phrase is 50 ) and average length of names in the table 30 - 50 symbols and 100 % to loading CPU:
    2 000 - 10 000 records in a second

    10000 records is clear time processing by code per second
    7 sec per 10000 records for use in real view with use of UDF and sorting

    all tested on low pefomance server (1proc *Intel PIII 900 MZ)

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts