VERIFICATION AND VALIDATION WITH RIPPLE-DOWN RULES

Verification to ensure a system's consistency and validation to meet t he user's criteria are essential elements in developing knowledge-base d systems for real world use. The normal practice is that there will b e initial knowledge acquisition attempting to build a complete system which will (should) then be verified and validated, There may be a cyc le through these steps till the system is complete, Maintenance is see n as a minor problem requiring the occasional repetition of the three stage process. The implicit assumption is that an expert has complete knowledge and that by a suitable knowledge acquisition process this is acquired. In fact, it seems rather than experts are incapable of reco unting how they reach a conclusion. Rather, when asked a question they justify that their conclusion is correct and their justification is t ailored to the specific context of the inquiry, Experts are best at ju stifying why one conclusion is to be preferred over another. This lead s to a knowledge acquisition methodology, Ripple-down Rules, in which the knowledge base undergoes on-going development based on correcting errors. Each new correction or justification is considered only in the context of the same mistake being made, The method also constrains th e expert's choices to ensure that any new knowledge added is valid whi le the knowledge base structure ensures the knowledge is verified. Ver ification and validation are not separate tasks, but constraints on kn owledge acquisition which itself continues throughout the life of the system. This provides a closer match with the normal evolution of huma n knowledge and expertise. The overall approach has itself been valida ted by the development of a large medical expert system and through si mulation studies, The medical system has been developed while in routi ne use and has only involved experts without any knowledge engineering support or skill in its development. (C) 1996 Academic Press Limited