1.3. O U T L I N E OF THIS B O O K 7 approaches and argues for the use of probability theory. 2 We will not present that argument here. Rather, we accept probability theory as being the way to handle uncertainty and explain why we choose to describe informatics algo- rithms that use the model-based probabilistic approach. A h e u r i s t i c a l g o r i t h m uses a commonsense rule to solve a problem. Ordi- narily, heuristic algorithms have no theoretical basis and therefore do not enable us to prove results based on assumptions concerning a system. An example of a heuristic algorithm is the one developed for collaborative filtering in Chapter 11, Section 11.1. An a b s t r a c t m o d e l is a theoretical construct that represents a physical process with a set of variables and a set of quantitative relationships (axioms) among them. We use models so we can reason within an idealized framework and thereby make predictions/determinations about a system. We can math- ematically prove these predictions/determinations are "correct," but they are correct only to the extent that the model accurately represents the system. A m o d e l - b a s e d a l g o r i t h m therefore makes predictions/determinations within the framework of some model. Algorithms that make predictions/determinations within the framework of probability theory are model-based algorithms. We can prove results concerning these algorithms based on the axioms of probability theory, which are discussed in Chapter 2. We concentrate on such algorithms in this book. In particular, we present algorithms that use Bayesian networks