Basic Concepts of Expert System Shells and an Efficient Model for Knowledge Acquisition

employs knowledge and inference procedures to solve problems that are considered difficult enough to require significant human expertise for their solutions [12]. Since the implementation of the first expert systems (the DENDRAL, MYCIN, Dipmeter Advisor, XCON, to mention but a few.) in the early 1970s, and the successes that followed, coupled with the advantages that humankind benefited from these early expert systems and their successors; one would expect that four decades later, there would have been a boom in the use of expert systems to perform more specialized tasks in different specific domains and application areas. The reasons for this could stem, asides other non-intrinsically-technical factors (such as financial implication, cultural and religious beliefs and restrictions surrounding the implementation of Expert Systems especially in third-world African Countries), either from a lack of adequate knowledge of the existence of Expert System Shells and/or the facilities, utilities and tools they provide that make the development of new expert systems sort of a “pretty easy endeavour”. This paper lays emphasis on the basic concept of Expert System Shells with the aim of highlighting their basic usage and creating some form of in-depth generic exposé into some of the component facilities and utilities they provide; proposing a more efficient model for Knowledge Acquisition – which would make the Knowledge Acquisition Bottleneck of developing Expert Systems a thing of no more concern – as well as highlight the functions performed by Expert System Shells, in a bid to emphasize their reusability across different other specific application areas either within or outside the same domain