Re: How would Bayes Network encode disjunctive and conjunctive terms?

From: Harris Georgiou (xgeorgiou@yahoo.com)
Date: Wed Feb 06 2002 - 13:55:33 MET

  • Next message: Sanya Mitaim: "CFP: IEEE ICIT 02"

    "Moses Chan" <moses_w_chan@raytheon.com> wrote in message
    news:e2175f.0201251517.55b6e409@posting.google.com...
    > Hi all,
    >
    > Could someone comment on how would Bayesian Network encode these rules:
    >
    > 1) if A then (B and C);
    > 2) if A then (B or C);
    > 3) if (A and B) and C;
    > 4) if (A or B) then C.
    >
    > Specifically, how would one determine the difference between conjunctive
    and
    > disjunctive term using conditional independency?

    In general, for solid Bayesian networks you should use individual
    probabilities for every joined fact. If this is not possible, you could use
    joined probabilities for joined conditional facts. For joined results
    (especially "AND") the only thing you can do is combine any previous
    (statistical) knowledge about A, B, C and produce some estimation on joined
    facts using inclusion-exclusion theory from sets. For example, if B and C
    are statistically independent, then P{(B.and.C)|A) should be equal to
    P(B|A)*P(C|A).

    --
    

    Harris

    - 'Malo e lelei ki he pongipongi!'

    ############################################################################ This message was posted through the fuzzy mailing list. (1) To subscribe to this mailing list, send a message body of "SUB FUZZY-MAIL myFirstName mySurname" to listproc@dbai.tuwien.ac.at (2) To unsubscribe from this mailing list, send a message body of "UNSUB FUZZY-MAIL" or "UNSUB FUZZY-MAIL yoursubscription@email.address.com" to listproc@dbai.tuwien.ac.at (3) To reach the human who maintains the list, send mail to fuzzy-owner@dbai.tuwien.ac.at (4) WWW access and other information on Fuzzy Sets and Logic see http://www.dbai.tuwien.ac.at/ftp/mlowner/fuzzy-mail.info (5) WWW archive: http://www.dbai.tuwien.ac.at/marchives/fuzzy-mail/index.html



    This archive was generated by hypermail 2b30 : Wed Feb 06 2002 - 13:59:13 MET