Contemporary Artificial Intelligence

Chapter 1 1

Learning Determ inistic

Models

In Part I (Logical Intelligence), we solved problems using models that repre-

sen t ed deterministic relationships. For example, in Section 2.3.1, we introduced

decision trees, and we showed how we could use a decision tree to determine the

family of a plan t from its properties. In Part II (Probabilistic Intelligence), w e

solved problems using models that represented probabilistic relationships. For

example, in Section 6.7, we introduced Promedas, which produces a differential

(probabilistic) diagnosis using a set of patient findings such as history data,

physical findings, and laboratory data. In both parts, ...