[UAI] Re: Fuzzy sets vs. Bayesian Networks

From: C.Rajagopalan (raja@igcar.ernet.in)
Date: Fri Feb 25 2000 - 09:18:52 PST

  • Next message: Scott Ferson: "Re: [UAI] Fuzzy sets vs. Bayesian Network"

    Dear Members:

    >From a practical point of view, I feel that fuzzy sets would "solve" a given
    decision support problem, given the fact that we always do not have enough
    data.

    That being said, if we use a fuzzy approach for uncertainty management,

    a) for a given problem, is there a unique way to represent the membership
    functions ?

    b) how to ascertain that this indeed >>is<< the membership function for a
    given set ?

    c) what are the "qualifications" of an expert who recommends such and such
    membership functions are appropriate ?

    Sincerely,

    C.Rajagopalan,
    Division for PIE and NDT Development,
    Indira Gandhi Centre for Atomic Research,
    Kalpakkam 603 102 TamilNadu
    INDIA

    E-Mail : raja@igcar.ernet.in
    Phone : +91 4114 40208
    Fax : +91 4114 40356
    Web : http://deskpack.webjump.com/index.html
                 http://members.tripod.com/DESKPACK/



    This archive was generated by hypermail 2b29 : Fri Feb 25 2000 - 09:20:30 PST