Misuse and nonuse of Knowledge-Based Systems:The past experiences revisited

Abstract

It is difficult to determine the number of expert systems or Knowledge-Based Systems (KBSs) that really are operational within companies or administrations. It seems that a large number of such systems have never been used in operations and a rich literature stresses this point. We think that most of these references do not address some dimensions that are of paramount importance. In this paper, we provide a review of the literature according to three neglected dimensions: (1) the differences between automatic KBSs and non-automatic KBSs; (2) the types and importance of decisions that are involved; and (3) the types of data acquisition that is required. Keeping in mind these three dimensions, we review the literature about the acceptance of KBS and their use and point out that most of the observations can be interpreted along these three dimensions. Furthermore, these dimensions permit us to explain some failures and difficulties that have already been pointed out in several scientific domains in which interactivity is crucial, such as Decision Support Systems (DSSs). Our approach brings some new insights on the problems of KBS acceptance and leads us to propose some recommendations.