Knowledge elicitation as an inquiring system: towards a 'subjective' knowledge elicitation methodology

Many researchers in the field of knowledge elicitation (KE) have recognized the importance of the role of ‘expertise’ for the development of expert systems. This recognition has led to attempts to classify knowledge ‘types’ and define ‘genericxyr task structures so as to reduce the complexity of the KE process (Gammack & Young, 1985; Breuker & Wielinga, 1987; Welbank, 1987). The success of such an approach has been debatable as expertise has proved difficult to classify (Burton et al., 1987), with different researchers adopting different catagorizations. The research reported in this paper attempts to develop a broad view of ‘expertise’ by drawing upon Vicker's notion of ‘appreciation’ (1965). This, it is argued, can help to provide a useful description of the way in which expertise is gained and practised. The model suggested also takes into account the more ‘subjective’ aspects of human expertise (i.e. it allows for such components as intuition, personal experience and belief, bias and even prejudice). The view of expertise put forward is developed as a theoretical basis for a practical subjective approach to KE. As a result of this model, the author suggests that instead of concentrating upon the process of extraction, KE may be more usefully thought of as a process of inquiry through which, as described by Churchman (1971), knowledge, or expertise is produced. Some ideas that deal with ‘inquiring systems’ are examined here as an appropriate framework for the KE process.