Knowledge elicitation and integration from multiple experts

The idea of this paper is based upon the belief that each person has his (or her) own habitual domain, in which he (or she) used to solve problems by his (or her) own way. If a person's habitual domain can not cover the problem to be solved, he (or she) will feel diffident and uneasy, and incorrect decisions may be made. For an expert to solve a problem within his (or her) habitual domain, overconfidence and ignorance may sometimes lead to mistakes. To cope with these problems, a knowledge acquisition method for multiple experts, MERGE, is proposed. Some experimental results on a practical application have shown that MERGE is able to enrich the elicited knowledge and improve the performance of the constructed expert systems by integrating expertise from multiple experts.<<ETX>>