GTKAT: a grounded theory based knowledge acquisition tool for expert systems

Abstract

Knowledge acquisition is a crucial problem area in the process of building expert systems. It has been identified as the knowledge engineering bottleneck. To improve the knowledge engineering approach, many lessons could be learned from the grounded theory. Grounded theory provides the knowledge engineer with an important tool for obtaining a faithful reflection of qualitative data derived from human experts and other sources. The aim of this work is to present the design of a knowledge acquisition tool based on the grounded theory methodology. It is argued that the grounded theory based knowledge acquisition tool has strong theoretical foundations and is capable of dealing with the unstructured interview data. It is also shown that there are valuable lessons in the transfer of techniques from social sciences to artificial intelligence (AI) research. >