Knowledge-Based Systems in the Commercial Environment

Knowledge-based systems are among the first applications of artificial intelligence to make the crossover from the laboratory to the real-world commercial environment. Typically, artificial intelligence systems have been implemented in the LISP programming language on specialized hardware. The experimental nature of early systems has allowed many of them the luxury of having little or no interface to existing hardware, software, or data. In this paper, arguments are presented to demonstrate the feasibility of implementing knowledge-based shell systems using traditional hardware and software. Also, an architecture is proposed for knowledge-based shell systems that is compatible with the software development environment of large commercial information systems organizations. To demonstrate these concepts, an example system is shown.