Knowledge Acquisition from User Programs by Computer with Own Knowledge-Based System

Abstract

This paper deals with the problem of knowledge acquisition from user programs for the knowledge base of the computer. Various forms of knowledge included in user programs are considered. First, knowledge expressed in the form of rules included in a user program is discussed. It is shown that this knowledge, extracted from the user program, can be transformed into a symbolic expression. This symbolic expression is compared with the symbolic expression that represents the knowledge contained in a knowledge base. Such elements are then extracted from the symbolic expression which, as acquired knowledge, should be added to the knowledge base. Moreover, knowledge expressed as an algorithm for the solution of problems is considered. It is shown that such knowledge is contained both in the structure and in the sequence of operations of user programs. These parts of user programs are transformed into symbolic expressions. On the basis of these expressions the knowledge acquisition process is realized. The method of knowledge acquisition, presented in this paper, can be used in decision support systems.