Search All of the Math Forum:
Views expressed in these public forums are not endorsed by
NCTM or The Math Forum.



how to measure what event gives the most helpfull information ?
Posted:
Jul 31, 1996 7:07 AM


Hi,
I have the following problem.
I have a list of hypotheses, A_1, A_n. The Hypotheses are exclusive, only one may be satisfied at the some time. I have a list of evidences for the hypotheses, i.e. other events, let they be X_1, .., X_m, that influence the hypotheses.
I have a rought measure of how events influence facts: from a formal point of view I have something as P(A_j,X_i) for each i, j. I would like to select the events that give the widest information about the hypotheses, i.e. the events that are most helpfull in orders to decide what hypothesis is the most likely.
Could anybody gives me some hints ?
Franco



