Results 1 to 2 of 2

Thread: Best Starter Algorithm for New Data Mining Initiative?

  1. 04-10-2007, 05:57 AM #1
    Chase_Idol
    Chase_Idol is offline Registered User
    Join Date
    Mar 2007
    Posts
    6

    Best Starter Algorithm for New Data Mining Initiative?

    I see that several algorithms are included with the MSSQL Server 2005 Data Mining component. What's the best approach - and algorithm choice - for "kick starting" our new initiative?

    Any tips appreciated!
    Reply With Quote Reply With Quote
  2. 04-11-2007, 09:14 AM #2
    Island1
    Island1 is offline DBJ Author
    Join Date
    Jun 2004
    Location
    Atlanta and Manhattan
    Posts
    607

    Naive Bayes is my Suggestion ...

    I typically find the best "starter" for a new mining initiative to be a Naive Bayes model, for several reasons. It's a good introduction to several concepts in one algorithm. Chiefly, if you configure it for "input" and "predictable" on all (nonkey, of course) columns, then you can get a good "discovery" experience, whereby you will likely learn more about your data up front. This obviously helps in crafting subsequent models, even with additional algorithms...

    Good Luck.

    Bill
    Reply With Quote Reply With Quote
« Previous Thread | Next Thread »

Posting Permissions

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

Forum Rules