[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Applicable Models for Trust Calculations (Re: Dempster-Shafer Theory and Belief Networks (Re: Transitive trust))



At 09:51 PM 5/7/96 -0700, Timothy C. May wrote:

>In particular, one form of belief representation seems especially relevant:
>Dempster-Shafer theory.

Two quick things: (more in my forthcoming thesis)

1. It is very instructive to consider systems for considering trust and
belief (and Dempster-Shafer is a rather nice one). However, these systems
(my own decision analysis discussion to a degree, but less so with respect
to information gathering) and Dempster-Shafer in particular require the
events to be independent:

"Further, Dempster-Shafer theory provides rules for combining probabilities
and thus for propagating measures through the system. This fourth point is
possibly the most attractive, but it is also one of the most controversial
since the propagation method is an extension of the multiplication rule for
independent events. Since many of the applications involve events that are
surely dependent, that rule is, by classical statistical criteria,
inapplicable. The tendency to assume that events are independent unless
proven otherwise has stimulated a large proportion of the criticism of
probability approaches; as it stands, Dempster-Shafer theory suffers the
same ill" [Shapiro (ed.) Encyclopedia of Aritificial Intelligence.
/Reasoning, Default./ p 846.]

The very term "Web of Trust" makes one pause with respect to independent events.

2. Just as a comment, it seems there are different meaning of "transitiviy"
in related but different disciplines, for instance in value and utility
functions, one requires transitivity, but in a way different from what Hal
discusses:

"For any three possible set os of consequences, X1, X2, and X3, if X1 > X2
and X2 > X3 then the preference is transitive such that X1 > X3."

[deNeufville, R. Applied Systems Analysis: Engineering Planning and
Technology Management, 1990, p 313.]

This refers to a single person's preferences (rather than 3 individuals in a
"network.")
_______________________
Regards,       Men govern nothing with more difficulty than their tongues,
               and can moderate their desires more than their words. -Spinoza
Joseph  Reagle      http://farnsworth.mit.edu/~reagle/home.html
[email protected]      E0 D5 B2 05 B6 12 DA 65  BE 4D E3 C1 6A 66 25 4E