ON GENERAL AND EXPERT KNOWLEDGE-BASED METHODS IN PROBLEM-SOLVIN

There are two distinct parts in this dissertation. The first part deals with general problem-solving methods, the second one with expert knowledge-based systems.;In the first part, an alternative view to Newell and Simon theory of human problem-solving is proposed. The main thesis is that the solution of a problem consists of: (1) the identification of its underlying structure, and (2) the invocation of the appropriate heuristics for that underlying structure. Essential to the proposed theory is that the same underlying structure permeates a large class of problems. Hence the same heuristics may be used to solve a large class of problems ranging from algebra problems to the Missionaries-and-Cannibals task. The notion of heuristics that I have proposed is a primitive construct wired in the human cognitive organization. I have called it proximity or closeness function. By using the notion of proximity defined on an underlying structure, one of the connotations of the notion of reformulation has been precisely defined. A class of problems that has been used as testbed for Artificial Intelligence ideas on problem-solving has been solved using the notion of proximity and reformulation.;The second part deals with medical diagnosis. A diagnostician, when he/she arrives at a diagnosis or diagnoses, has invoked some concepts. They can be diseases, causes of them, or other notions that are relevant to the diagnosis. These concepts form a hierarchical structure similar to a botanical classification. The diagnostician's knowledge is distributed through this hierarchy. The concepts in the hierarchy provide the criteria to organize under them small pieces of knowledge represented in the form of production rules. They extend the capabilities of production rules to more complex problem-solving situations. The rules under each concept are further organized into three subgroups: exclusionary, confirmatory and recommendation rules. During problem solving, the concepts are taken as specialists. They interact and communicate among themselves by means of a blackboard.