Re: [UAI] About Data Analysis

From: Marek J. Druzdzel (marek@andrew.cmu.edu)
Date: Sat Mar 18 2000 - 19:02:59 PST

  • Next message: Nathalie Jitnah: "[UAI] PhD thesis announcement"

    --On Saturday, March 18, 2000, 4:59 PM -0800 JosX Carlos
    <jrocha@convoy.com.br> wrote:

    > I'm help a friend who's working in analysis
    > of business data. The data was acquired from a questionnaire.
    > Each question has a set of answers and the the user marks
    > one of that. Futhermore, the answers are made in some way
    > that the second involves the first, third involves the
    > second, and so on.
    > I wold like to know if is correct treat this data
    > how mutually exclusive. If it's incorrect could any one
    > gimme some suggestion.

    JosX,

    Assuming that each data record is collected from a different individual, I
    don't see any reason for not assuming that they are independent.

    As far as answers to individual questions are concerned, the answers seem
    to be dependent. When learning a Bayesian network model for these data,
    you can model this dependence as a-priori knowledge (forced arcs).
    Hope this helps,

    Marek
    --------------------------------------------------------------------------
    Marek J. Druzdzel http://www.pitt.edu/~druzdzel



    This archive was generated by hypermail 2b29 : Sat Mar 18 2000 - 19:07:08 PST